On 09/12/18 06:03 AM, Carlos E. R. wrote:

But since I upgraded to 15.0 a few weeks ago, I got this problem. And it is firefox the culprit.

I am running 42.3 and had similar thoughts. Now I'm not running with Dolphin and while things are better, it still freezes, but not nearly so much, and yes, killing FF still seems to release it. Elsewhere I've read that there is a problem with the 4.19 kernel. Something to do with the networking component, but that may affect unix-sockets as well as internet connections. Supposedly this gets fixed in 4.20. Roll on 4.20 (there are tools to look at what unix sockets are active) (netstat, ss -xp, fuser, and especially 'lsof' because it can examine caching too) I'm still running FF 52.8 and was running that back in the 4.17 days when this wasn't a problem. Honestly though, I think this is a kernel problem not a FF problem. Certainly if it happens with Chrome instead of FF and with Dolphin then what to they have in common? -- A: Yes. > Q: Are you sure? >> A: Because it reverses the logical flow of conversation. >>> Q: Why is top posting frowned upon? -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 10/12/18 07:56 AM, Carlos E. R. wrote:

It is a kernel problem, yes. Something in FF causes the kernel to go berseck into a loop of (guess) load-unload memory to swap.

And mind, swapping in my machines is fast, it is on SSD.

let's not forget that you do have control over swap, though it is somewhat of an arcane art. You can control the 'swapiness' in various ways: - how little memory triggers swap; - how much memory must be recovered in one go oh my, all that under /proc/sys/vm! https://www.youtube.com/watch?v=9iJ06dbZ3fw sync; echo 3 | sudo tee /proc/sys/vm/drop_caches https://www.youtube.com/watch?v=f0pNgK2Em-M (about setting 'swapiness') (is there a TL;DR for video? this could have been done in a page or two of text, a faster read ) -- A: Yes. > Q: Are you sure? >> A: Because it reverses the logical flow of conversation. >>> Q: Why is top posting frowned upon? -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 10/12/2018 14.29, jdd@dodin.org wrote:

Le 10/12/2018 à 14:21, Anton Aylward a écrit :

...

...

And mind, swapping in my machines is fast, it is on SSD.

just as a wild guess, if it's the machine where you did the tests, didn't you forget an odd setup used then?

Like what? The machine is of course upgraded from older versions, and the entire home is the same, obviously. -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)

On 09/12/2018 13:03, Carlos E. R. wrote:

Something in firefox causes swap to trash that did not before, with the same amount of tabs or even more.

Have a look at this bug: https://bugzilla.opensuse.org/show_bug.cgi?id=1114838 and the mozilla bug it's linked to. Sorry if I'm interfering, I haven't had time to read the whole thread. Dave P -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Carlos E. R. wrote:

On 09/12/2018 10.17, David C. Rankin wrote:

...

On 12/08/2018 03:15 PM, Carlos E. R. wrote:

...

Hi,

Sometimes since I upgraded my laptop to 15.0, and after it has been sitting alone doing nothing for hours in the sitting room, I get to it and it doesn't respond. Right now I tried to connect to it via ssh and it is responding very slow - yast failed to start, for instance.

Carlos,

I had something similar happen 2 days ago on my Leap 15 install. It had been up a couple of days and I went back into the desktop where I had an open xterm and no commands would work at all. It was like the PATH had disappeared.

From memory it was like:

$ ssh nirvana /bin/ssh unknown command

No, this is different.

The "kswapd0" kernel process was stuck at 25% CPU, so something was trashing like hell, in a loop. Thus starting any process took minutes. I was lucky to even be able to ssh in and finally kill firefox.

As a work around, maybe use that trick with a cgroup, it was discussed here a while back. -- Per Jessen, Zürich (10.4°C) http://www.dns24.ch/ - your free DNS host, made in Switzerland. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Carlos E. R. wrote:

On 09/12/2018 13.33, Per Jessen wrote:

...

Carlos E. R. wrote:

...

On 09/12/2018 10.17, David C. Rankin wrote:

...

On 12/08/2018 03:15 PM, Carlos E. R. wrote:

...

Hi,

Sometimes since I upgraded my laptop to 15.0, and after it has been sitting alone doing nothing for hours in the sitting room, I get to it and it doesn't respond. Right now I tried to connect to it via ssh and it is responding very slow - yast failed to start, for instance.

Carlos,

I had something similar happen 2 days ago on my Leap 15 install. It had been up a couple of days and I went back into the desktop where I had an open xterm and no commands would work at all. It was like the PATH had disappeared.

From memory it was like:

$ ssh nirvana /bin/ssh unknown command

No, this is different.

The "kswapd0" kernel process was stuck at 25% CPU, so something was trashing like hell, in a loop. Thus starting any process took minutes. I was lucky to even be able to ssh in and finally kill firefox.

As a work around, maybe use that trick with a cgroup, it was discussed here a while back.

Ok, but doing what with it? I did not see any value that was that abnormal. I mean, not with firefox. I can not touch kswapd0, that's kernel.

I think exactly the same thing as was discussed, make sure there is always memory left for the GUI to be responsive. -- Per Jessen, Zürich (8.9°C) http://www.hostsuisse.com/ - dedicated server rental in Switzerland. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Carlos E. R. wrote:

On 09/12/2018 16.42, Per Jessen wrote:

...

Carlos E. R. wrote:

...

On 09/12/2018 13.33, Per Jessen wrote:

...

Carlos E. R. wrote:

...

On 09/12/2018 10.17, David C. Rankin wrote:

...

On 12/08/2018 03:15 PM, Carlos E. R. wrote: > Hi, > > Sometimes since I upgraded my laptop to 15.0, and after it has > been sitting alone doing nothing for hours in the sitting room, > I get to it and it doesn't respond. Right now I tried to connect > to it via ssh and it is responding very slow - yast failed to > start, for instance. >

Carlos,

I had something similar happen 2 days ago on my Leap 15 install. It had been up a couple of days and I went back into the desktop where I had an open xterm and no commands would work at all. It was like the PATH had disappeared.

From memory it was like:

$ ssh nirvana /bin/ssh unknown command

No, this is different.

The "kswapd0" kernel process was stuck at 25% CPU, so something was trashing like hell, in a loop. Thus starting any process took minutes. I was lucky to even be able to ssh in and finally kill firefox.

As a work around, maybe use that trick with a cgroup, it was discussed here a while back.

Ok, but doing what with it? I did not see any value that was that abnormal. I mean, not with firefox. I can not touch kswapd0, that's kernel.

I think exactly the same thing as was discussed, make sure there is always memory left for the GUI to be responsive.

I'm normally running my machines with much less memory and more swap than that.

This instant:

...

KiB Mem: 8174456 total, 6002948 used, 2171508 free, 81904 buffers KiB Swap: 25165820 total, 5668656 used, 19497164 free. 2059292 cached Mem

And there is no issue. What happened on that laptop was something else.

Notice that CLI over ssh was not responsive. Nothing was responsive, I can not give more resources to the GUI to solve it.

Notice I didn't say "give more resources", I suggested you reserve memory to make sure the GUI remains responsive.

Rather, it would be limiting some resource to FF, but which?

Memory. By reserving memory for the GUI, you are also limiting what other app can use. -- Per Jessen, Zürich (5.7°C) http://www.dns24.ch/ - free dynamic DNS, made in Switzerland. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Carlos E. R. wrote:

On 10/12/2018 14.52, Per Jessen wrote:

...

Carlos E. R. wrote:

...

On 09/12/2018 16.42, Per Jessen wrote:

...

Carlos E. R. wrote:

...

On 09/12/2018 13.33, Per Jessen wrote:

...

Carlos E. R. wrote:

> On 09/12/2018 10.17, David C. Rankin wrote: >> On 12/08/2018 03:15 PM, Carlos E. R. wrote: >>> Hi, >>> >>> Sometimes since I upgraded my laptop to 15.0, and after it has >>> been sitting alone doing nothing for hours in the sitting >>> room, I get to it and it doesn't respond. Right now I tried to >>> connect to it via ssh and it is responding very slow - yast >>> failed to start, for instance. >>> >> >> Carlos, >> >> I had something similar happen 2 days ago on my Leap 15 >> install. It had been >> up a couple of days and I went back into the desktop where I >> had an open xterm and no commands would work at all. It was >> like the PATH had disappeared. >> >> From memory it was like: >> >> $ ssh nirvana >> /bin/ssh unknown command > > No, this is different. > > The "kswapd0" kernel process was stuck at 25% CPU, so something > was trashing like hell, in a loop. Thus starting any process > took minutes. I was lucky to even be able to ssh in and finally > kill firefox.

As a work around, maybe use that trick with a cgroup, it was discussed here a while back.

Ok, but doing what with it? I did not see any value that was that abnormal. I mean, not with firefox. I can not touch kswapd0, that's kernel.

I think exactly the same thing as was discussed, make sure there is always memory left for the GUI to be responsive.

I'm normally running my machines with much less memory and more swap than that.

This instant:

...

KiB Mem: 8174456 total, 6002948 used, 2171508 free, 81904 buffers KiB Swap: 25165820 total, 5668656 used, 19497164 free. 2059292 cached Mem

And there is no issue. What happened on that laptop was something else.

Notice that CLI over ssh was not responsive. Nothing was responsive, I can not give more resources to the GUI to solve it.

Notice I didn't say "give more resources", I suggested you reserve memory to make sure the GUI remains responsive.

...

Rather, it would be limiting some resource to FF, but which?

Memory. By reserving memory for the GUI, you are also limiting what other app can use.

I don't think that will help. Notice that the CLI was not responsive. Nothing was responsive. Giving more memory to the GUI (what process) will not help with the CLI.

After you start the GUI, all subsequent processes are in the same cgroup. That way you will have e.g. konsole remaining responsive. I think we ought to have this config by default, at least on machines with 4Gb or more. -- Per Jessen, Zürich (4.8°C) http://www.hostsuisse.com/ - virtual servers, made in Switzerland. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Carlos E. R. wrote:

On 10/12/2018 15.04, Per Jessen wrote:

...

Carlos E. R. wrote:

...

On 10/12/2018 14.52, Per Jessen wrote:

...

Carlos E. R. wrote:

...

...

...

...

Notice I didn't say "give more resources", I suggested you reserve memory to make sure the GUI remains responsive.

...

Rather, it would be limiting some resource to FF, but which?

Memory. By reserving memory for the GUI, you are also limiting what other app can use.

I don't think that will help. Notice that the CLI was not responsive. Nothing was responsive. Giving more memory to the GUI (what process) will not help with the CLI.

After you start the GUI, all subsequent processes are in the same cgroup. That way you will have e.g. konsole remaining responsive. I think we ought to have this config by default, at least on machines with 4Gb or more.

No, CLI over ssh is not on that group.

Correct, it isn't GUI but presumably you wouldn't need the remote access when the GUI remains responsive. -- Per Jessen, Zürich (5.9°C) http://www.cloudsuisse.com/ - your owncloud, hosted in Switzerland. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Carlos E. R. wrote:

I prefer to don't change anything yet till we really know what is happening, and for that I need to know what command to issue on the CLI that gives the same information as the sysrq-m that Andrei Borzenkov suggested.

It's the normal sysrq sequence - I think it only works in a console: Ctrl-Alt-Sysrq-m -- Per Jessen, Zürich (3.2°C) http://www.dns24.ch/ - your free DNS host, made in Switzerland. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Carlos E. R. wrote:

On 11/12/2018 10.11, Per Jessen wrote:

...

Carlos E. R. wrote:

...

I prefer to don't change anything yet till we really know what is happening, and for that I need to know what command to issue on the CLI that gives the same information as the sysrq-m that Andrei Borzenkov suggested.

It's the normal sysrq sequence - I think it only works in a console:

Ctrl-Alt-Sysrq-m

No, I need a command to do via ssh, keyboard does not respond.

echo m >/proc/sysrq-trigger -- Per Jessen, Zürich (3.8°C) http://www.dns24.ch/ - free dynamic DNS, made in Switzerland. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

09.12.2018 14:13, Carlos E. R. пишет:

The "kswapd0" kernel process was stuck at 25% CPU, so something was trashing like hell, in a loop. Thus starting any process took minutes. I was lucky to even be able to ssh in and finally kill firefox.

25% is likely one logical CPU (or core), i.e. kswapd consumes 100% of allocated CPU resources. kswapd is started (woken up) when memory becomes short and attempts to free some memory. So in your case it looks like it fails to free enough memory (at least for for long time) and is constantly started again. Every time it needs to traverse page list to find candidates to release so it is time consuming. When you stop firefox you free large amount of memory which returns system to normal state. It is impossible to say why it happened from top output alone. sysrq-m output may provide some clue.

10.12.2018 16:11, Carlos E. R. пишет:

...

It is impossible to say why it happened from top output alone. sysrq-m output may provide some clue.

I know, there is not enough information.

How can I do sysrq-m over ssh? Because locally the machine did not respond at all, black screen.

Output goes into dmesg so should be captured by journald/klogd/whatever. Also /proc/meminfo displays similar information.

http://www.linuxhowtos.org/Tips%20and%20Tricks/sysrq.htm

Alt+SysRQ+m prints memory information to the console.

Ok, so, is there a command to dump that information on the ssh session?

On 11/12/2018 12.07, Carlos E. R. wrote:

On 11/12/2018 05.13, Andrei Borzenkov wrote:

...

10.12.2018 16:11, Carlos E. R. пишет:

I will do "echo m >/proc/sysrq-trigger" now. I don't know where it will dump the information.

Ah, on syslog, which I have activated on ssh since yesterday. This is it: <4.5> 2018-12-11 11:35:10 minas-tirith dbus-daemon 1360 - - [system] Failed to activate service 'org.freedesktop.nm_dispatcher': timed out (service_start_timeout=25000ms) <3.6> 2018-12-11 11:41:16 minas-tirith dbus-daemon 1360 - - [system] Activating via systemd: service name='org.freedesktop.nm_dispatcher' unit='dbus-org.freedesktop.nm-dispatcher.service' requested by ':1.76' (uid=0 pid=12743 comm="/usr/sbin/NetworkManager --no-daemon ") <4.5> 2018-12-11 11:41:34 minas-tirith dbus-daemon 1360 - - [system] Failed to activate service 'org.freedesktop.nm_dispatcher': timed out (service_start_timeout=25000ms) <3.6> 2018-12-11 12:01:40 minas-tirith dbus-daemon 1360 - - [system] Activating via systemd: service name='org.freedesktop.nm_dispatcher' unit='dbus-org.freedesktop.nm-dispatcher.service' requested by ':1.76' (uid=0 pid=12743 comm="/usr/sbin/NetworkManager --no-daemon ") <3.6> 2018-12-11 12:01:43 minas-tirith smartd 1358 - - Device: /dev/sda [SAT], SMART Usage Attribute: 194 Temperature_Celsius changed from 46 to 45 <4.5> 2018-12-11 12:01:59 minas-tirith dbus-daemon 1360 - - [system] Failed to activate service 'org.freedesktop.nm_dispatcher': timed out (service_start_timeout=25000ms) <0.6> 2018-12-11 12:06:40 minas-tirith kernel - - - [193727.277594] sysrq: SysRq : Show Memory <0.4> 2018-12-11 12:06:41 minas-tirith kernel - - - [193727.277612] Mem-Info: <0.4> 2018-12-11 12:06:41 minas-tirith kernel - - - [193727.277619] active_anon:736954 inactive_anon:147439 isolated_anon:0 <0.4> 2018-12-11 12:06:43 minas-tirith kernel - - - [193727.277619] active_file:1457 inactive_file:6905 isolated_file:32 <0.4> 2018-12-11 12:06:43 minas-tirith kernel - - - [193727.277619] unevictable:20 dirty:27 writeback:0 unstable:0 <0.4> 2018-12-11 12:06:45 minas-tirith kernel - - - [193727.277619] slab_reclaimable:10691 slab_unreclaimable:16620 <0.4> 2018-12-11 12:06:46 minas-tirith kernel - - - [193727.277619] mapped:22456 shmem:61655 pagetables:19913 bounce:0 <0.4> 2018-12-11 12:06:48 minas-tirith kernel - - - [193727.277619] free:25722 free_pcp:615 free_cma:0 <0.4> 2018-12-11 12:06:49 minas-tirith kernel - - - [193727.277624] Node 0 active_anon:2947816kB inactive_anon:589756kB active_file:5828kB inactive_file:27620kB unevictable:80kB isolated(anon):0kB isolated(file):128kB mapped:89824kB dirty:108kB writeback:0kB shmem:246620kB shmem_thp: 0kB shmem_pmdmapped: 0kB anon_thp: 399360kB writeback_tmp:0kB unstable:0kB all_unreclaimable? no <0.4> 2018-12-11 12:06:50 minas-tirith kernel - - - [193727.277625] Node 0 DMA free:15504kB min:272kB low:340kB high:408kB active_anon:396kB inactive_anon:0kB active_file:0kB inactive_file:0kB unevictable:0kB writepending:0kB present:15988kB managed:15904kB mlocked:0kB slab_reclaimable:4kB slab_unreclaimable:0kB kernel_stack:0kB pagetables:0kB bounce:0kB free_pcp:0kB local_pcp:0kB free_cma:0kB <0.4> 2018-12-11 12:06:52 minas-tirith kernel - - - [193727.277631] lowmem_reserve[]: 0 2854 3813 3813 3813 <0.4> 2018-12-11 12:06:53 minas-tirith kernel - - - [193727.277635] Node 0 DMA32 free:66372kB min:50384kB low:62980kB high:75576kB active_anon:2295568kB inactive_anon:403584kB active_file:3220kB inactive_file:22580kB unevictable:80kB writepending:88kB present:3030336kB managed:2936620kB mlocked:80kB slab_reclaimable:22328kB slab_unreclaimable:31900kB kernel_stack:9472kB pagetables:61640kB bounce:0kB free_pcp:1300kB local_pcp:1052kB free_cma:0kB <0.4> 2018-12-11 12:06:54 minas-tirith kernel - - - [193727.277641] lowmem_reserve[]: 0 0 958 958 958 <0.4> 2018-12-11 12:06:54 minas-tirith kernel - - - [193727.277645] Node 0 Normal free:21012kB min:16920kB low:21148kB high:25376kB active_anon:651852kB inactive_anon:186172kB active_file:2744kB inactive_file:4908kB unevictable:0kB writepending:20kB present:1048576kB managed:981716kB mlocked:0kB slab_reclaimable:20432kB slab_unreclaimable:34580kB kernel_stack:4272kB pagetables:18012kB bounce:0kB free_pcp:1160kB local_pcp:112kB free_cma:0kB <0.4> 2018-12-11 12:06:55 minas-tirith kernel - - - [193727.277651] lowmem_reserve[]: 0 0 0 0 0 <0.4> 2018-12-11 12:06:56 minas-tirith kernel - - - [193727.277654] Node 0 DMA: 2*4kB (ME) 1*8kB (E) 2*16kB (ME) 3*32kB (UME) 4*64kB (UME) 2*128kB (UE) 2*256kB (UE) 2*512kB (ME) 3*1024kB (UME) 1*2048kB (E) 2*4096kB (M) = 15504kB <0.4> 2018-12-11 12:06:57 minas-tirith kernel - - - [193727.277671] Node 0 DMA32: 464*4kB (UME) 1026*8kB (UE) 1063*16kB (UME) 468*32kB (UME) 142*64kB (UME) 35*128kB (UM) 9*256kB (M) 9*512kB (M) 4*1024kB (M) 0*2048kB 0*4096kB = 66624kB <0.4> 2018-12-11 12:06:59 minas-tirith kernel - - - [193727.277686] Node 0 Normal: 4027*4kB (UMEH) 371*8kB (UMEH) 57*16kB (UMH) 14*32kB (MH) 7*64kB (H) 1*128kB (H) 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 21012kB <0.6> 2018-12-11 12:07:00 minas-tirith kernel - - - [193727.277700] Node 0 hugepages_total=0 hugepages_free=0 hugepages_surp=0 hugepages_size=2048kB <0.4> 2018-12-11 12:07:01 minas-tirith kernel - - - [193727.277701] 94422 total pagecache pages <0.4> 2018-12-11 12:07:02 minas-tirith kernel - - - [193727.277704] 24309 pages in swap cache <0.4> 2018-12-11 12:07:03 minas-tirith kernel - - - [193727.277706] Swap cache stats: add 626259, delete 602108, find 198766/294449 <0.4> 2018-12-11 12:07:04 minas-tirith kernel - - - [193727.277706] Free swap = 4814008kB <0.4> 2018-12-11 12:07:04 minas-tirith kernel - - - [193727.277707] Total swap = 6289412kB <3.6> 2018-12-11 12:07:05 minas-tirith dbus-daemon 1360 - - [system] Activating via systemd: service name='org.freedesktop.nm_dispatcher' unit='dbus-org.freedesktop.nm-dispatcher.service' requested by ':1.76' (uid=0 pid=12743 comm="/usr/sbin/NetworkManager --no-daemon ") <0.4> 2018-12-11 12:07:06 minas-tirith kernel - - - [193727.277708] 1023725 pages RAM <0.4> 2018-12-11 12:07:07 minas-tirith kernel - - - [193727.277709] 0 pages HighMem/MovableOnly <0.4> 2018-12-11 12:07:08 minas-tirith kernel - - - [193727.277709] 40165 pages reserved <0.4> 2018-12-11 12:07:09 minas-tirith kernel - - - [193727.277710] 0 pages hwpoisoned <4.5> 2018-12-11 12:07:34 minas-tirith dbus-daemon 1360 - - [system] Failed to activate service 'org.freedesktop.nm_dispatcher': timed out (service_start_timeout=25000ms) Is that all the info, or is it waiting for more? 'top' finally responded: top - 12:13:25 up 4 days, 10:12, 5 users, load average: 41.63, 44.29, 44.90 Tasks: 306 total, 1 running, 305 sleeping, 0 stopped, 0 zombie %Cpu(s): 0.1 us, 27.9 sy, 0.0 ni, 0.0 id, 71.3 wa, 0.0 hi, 0.7 si, 0.0 st KiB Mem : 3934240 total, 103460 free, 3510408 used, 320372 buff/cache KiB Swap: 6289412 total, 4794296 free, 1495116 used. 20348 avail Mem PID USER PR NI VIRT RES SHR SWAP S %CPU %MEM TIME+ COMMAND 38 root 20 0 0 0 0 0 S 23.42 0.000 93:38.60 kswapd0 12526 cer 20 0 1453444 137488 2816 14312 D 0.915 3.495 145:51.32 chrome 12670 cer 20 0 1703752 151644 5356 15124 D 0.815 3.854 104:17.03 chrome 12563 cer 20 0 2089092 63880 4524 243644 D 0.486 1.624 93:38.95 chrome 15952 cer 20 0 876624 118512 16556 21176 D 0.372 3.012 6:01.63 chrome 14060 cer 20 0 538600 4968 176 9852 D 0.257 0.126 140:13.85 chrome 1 root 20 0 173804 1988 44 772 D 0.229 0.051 0:24.74 systemd 7 root 20 0 0 0 0 0 S 0.143 0.000 0:33.90 ksoftirqd/0 3164 cer 20 0 9590388 864960 22632 33832 D 0.114 21.99 7:10.45 firefox 4319 cer 20 0 865652 4660 8 4540 D 0.100 0.118 1:26.74 contarcorreo 2703 root 20 0 454108 21588 8972 16268 D 0.086 0.549 39:59.53 X 26386 root 20 0 62424 2040 1004 0 R 0.086 0.052 0:02.56 top 4050 cer 20 0 896904 868 0 6340 S 0.218 0.022 0:34.38 kdeconnectd 4084 cer 20 0 662676 1804 52 4344 S 0.218 0.046 3:38.08 panel-6-pulseau 26081 root 20 0 0 0 0 0 S 0.218 0.000 0:03.18 kworker/1:1 26386 root 20 0 62424 1660 624 0 R 0.218 0.042 0:02.50 top 1 root 20 0 173804 1988 0 772 D 0.194 0.051 0:24.58 systemd 7 root 20 0 0 0 0 0 S 0.194 0.000 0:33.80 ksoftirqd/0 2703 root 20 0 454108 21588 8972 16268 D 0.194 0.549 39:59.47 X 2488 ntp 20 0 55032 184 0 580 D 0.169 0.005 0:24.98 ntpd 3005 root 20 0 188696 1692 0 69688 D 0.169 0.043 0:53.62 /usr/sbin/spamd 3429 root 20 0 294872 656 0 852 D 0.169 0.017 0:10.02 upowerd 4059 cer 20 0 200388 964 44 1948 D 0.169 0.025 3:22.55 panel-14-multil 4067 cer 20 0 555352 6576 308 2052 D 0.169 0.167 14:37.33 panel-18-weathe 4183 cer 39 19 0.250t 2788 0 4624 S 0.169 0.071 0:08.00 baloo_file Now I typed a '1' to see each core load. Waiting... -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)

On 11/12/2018 12.37, Carlos E. R. wrote:

On 11/12/2018 12.15, Carlos E. R. wrote:

...

On 11/12/2018 12.07, Carlos E. R. wrote:

...

Ok, I wait for suggestions on what to do next.

Ok, I killed thunderbird, and the machine became just a bit more responsive. Then I killed chrome, and it became responsive. So the problem does not seem firefox itself. Probably the kernel not managing swap properly, some race condition. top - 13:30:59 up 4 days, 11:29, 5 users, load average: 1.17, 20.99, 35.36 Tasks: 271 total, 1 running, 270 sleeping, 0 stopped, 0 zombie %Cpu0 : 1.5 us, 1.2 sy, 0.0 ni, 97.3 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st %Cpu1 : 1.8 us, 0.6 sy, 0.0 ni, 97.3 id, 0.0 wa, 0.0 hi, 0.3 si, 0.0 st KiB Mem : 3934240 total, 1309424 free, 2121088 used, 503728 buff/cache KiB Swap: 6289412 total, 5290168 free, 999244 used. 1451636 avail Mem PID USER PR NI VIRT RES SHR SWAP S %CPU %MEM TIME+ COMMAND 2703 root 20 0 449392 24704 11492 15652 S 0.896 0.628 40:07.51 X 3421 cer 20 0 41464 648 0 280 S 0.597 0.016 14:26.36 top 3948 cer 20 0 567540 3404 392 3912 S 0.597 0.087 11:07.09 xfwm4 4030 cer 20 0 875056 7192 2900 4028 S 0.597 0.183 18:49.33 gkrellm 26386 root 20 0 62424 1044 0 0 R 0.597 0.027 0:10.36 top Sorting by memory (RES): top - 13:31:23 up 4 days, 11:29, 5 users, load average: 0.85, 19.32, 34.43 Tasks: 271 total, 1 running, 270 sleeping, 0 stopped, 0 zombie %Cpu0 : 1.5 us, 1.3 sy, 0.0 ni, 97.1 id, 0.0 wa, 0.0 hi, 0.1 si, 0.0 st %Cpu1 : 1.5 us, 0.7 sy, 0.0 ni, 97.7 id, 0.0 wa, 0.0 hi, 0.1 si, 0.0 st KiB Mem : 3934240 total, 1306420 free, 2123600 used, 504220 buff/cache KiB Swap: 6289412 total, 5290168 free, 999244 used. 1449112 avail Mem PID USER PR NI VIRT RES SHR SWAP S %CPU %MEM TIME+ COMMAND 3256 cer 20 0 2543792 805272 15840 30120 S 0.000 20.47 0:53.40 Web Content 3342 cer 20 0 2336788 640792 37524 32844 S 0.000 16.29 1:24.18 Web Content 3164 cer 20 0 9487996 350420 61860 52756 S 0.439 8.907 7:32.39 firefox 3381 cer 20 0 1764236 77596 6964 32360 S 0.000 1.972 0:19.15 Web Content 2703 root 20 0 449308 24744 11532 15652 S 1.023 0.629 40:07.73 X 3397 cer 20 0 1688972 20756 2840 44604 S 0.000 0.528 0:07.70 Web Content 4038 cer 20 0 1038948 15516 4120 18216 S 0.146 0.394 2:02.10 xfce4-terminal 4067 cer 20 0 555352 12760 6484 2044 S 0.439 0.324 14:42.96 panel-18-weathe 4003 cer 20 0 869880 11492 1344 6992 S 0.000 0.292 0:10.20 xfdesktop 4319 cer 20 0 865652 7916 3404 4680 S 0.000 0.201 1:33.80 contarcorreo 4269 cer 39 19 918436 7792 780 86756 S 0.000 0.198 0:34.50 tracker-miner-f 4030 cer 20 0 875056 7192 2900 4028 S 0.585 0.183 18:49.47 gkrellm 22832 root 20 0 18420 6700 2188 508 S 0.000 0.170 0:00.35 bash 3989 cer 20 0 747576 6016 3700 5944 S 0.000 0.153 0:16.09 xfce4-panel 3981 cer 20 0 1003288 5048 112 6780 S 0.000 0.128 0:07.42 Thunar This log excerpt finds when the problem started: <3.6> 2018-12-11 08:00:24 minas-tirith systemd 1 - - Started Timeline of Snapper Snapshots. <3.6> 2018-12-11 08:00:24 minas-tirith dbus-daemon 1360 - - [system] Activating service name='org.opensuse.Snapper' requested by ':1.284' (uid=0 pid=23787 comm="/usr/lib/snapper/systemd-helper --timeline ") (using servicehelper) <3.6> 2018-12-11 08:00:24 minas-tirith dbus-daemon 1360 - - [system] Successfully activated service 'org.opensuse.Snapper' <1.6> 2018-12-11 08:30:01 minas-tirith run-crons 24328 - - leafnode.cron: OK <1.6> 2018-12-11 08:30:01 minas-tirith run-crons 24328 - - mine-clocksync: OK <3.6> 2018-12-11 09:00:24 minas-tirith systemd 1 - - Started Timeline of Snapper Snapshots. <3.6> 2018-12-11 09:00:24 minas-tirith dbus-daemon 1360 - - [system] Activating service name='org.opensuse.Snapper' requested by ':1.286' (uid=0 pid=24967 comm="/usr/lib/snapper/systemd-helper --timeline ") (using servicehelper) <3.6> 2018-12-11 09:00:24 minas-tirith dbus-daemon 1360 - - [system] Successfully activated service 'org.opensuse.Snapper' <1.6> 2018-12-11 09:30:01 minas-tirith run-crons 25511 - - leafnode.cron: OK <1.6> 2018-12-11 09:30:01 minas-tirith run-crons 25511 - - mine-clocksync: OK <3.6> 2018-12-11 09:49:02 minas-tirith dbus-daemon 1360 - - [system] Activating via systemd: service name='org.freedesktop.nm_dispatcher' unit='dbus-org.freedesktop.nm-dispatcher.service' requested by ':1.76' (uid=0 pid=12743 comm="/usr/sbin/NetworkManager --no-daemon ") <4.5> 2018-12-11 09:49:28 minas-tirith dbus-daemon 1360 - - [system] Failed to activate service 'org.freedesktop.nm_dispatcher': timed out (service_start_timeout=25000ms) <3.6> 2018-12-11 10:01:41 minas-tirith smartd 1358 - - Device: /dev/sda [SAT], SMART Usage Attribute: 194 Temperature_Celsius changed from 62 to 50 <3.6> 2018-12-11 10:04:02 minas-tirith dbus-daemon 1360 - - [system] Activating via systemd: service name='org.freedesktop.nm_dispatcher' unit='dbus-org.freedesktop.nm-dispatcher.service' requested by ':1.76' (uid=0 pid=12743 comm="/usr/sbin/NetworkManager --no-daemon ") It starts after 'org.opensuse.Snapper' is activated, from 9:30 to 9:49. I don't have any btrfs partitions. -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)

On 11/12/2018 18.59, Liam Proven wrote:

On 11/12/2018 13:37, Carlos E. R. wrote:

...

It starts after 'org.opensuse.Snapper' is activated, from 9:30 to 9:49. I don't have any btrfs partitions.

So, er, remove it, then?

I've marked the btrfs tools as blacklisted on my box.

I can try. -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)

On 11/12/2018 15.37, Per Jessen wrote:

Carlos E. R. wrote:

...

...

Now I typed a '1' to see each core load. Waiting...

top - 12:18:07 up 4 days, 10:16, 5 users, load average: 41.77, 42.35, 43.84

That's quite a busy desktop you have there.

No, that's while it is crashed. I had been iddling for 12 hours, waiting for it to crash, doing nothing. This instant: top - 20:18:47 up 4 days, 18:17, 5 users, load average: 0.05, 0.06, 0.03 Tasks: 266 total, 1 running, 265 sleeping, 0 stopped, 0 zombie %Cpu0 : 1.5 us, 0.9 sy, 0.0 ni, 97.6 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st %Cpu1 : 1.2 us, 1.2 sy, 0.0 ni, 97.3 id, 0.0 wa, 0.0 hi, 0.3 si, 0.0 st KiB Mem : 3934240 total, 1089888 free, 2523072 used, 321280 buff/cache KiB Swap: 6289412 total, 5266360 free, 1023052 used. 1067920 avail Mem PID USER PR NI VIRT RES SHR SWAP S %CPU %MEM TIME+ COMMAND 3164 cer 20 0 9540212 685640 51092 52352 S 0.000 17.43 9:34.60 firefox 3342 cer 20 0 2336788 643512 34268 34168 S 0.000 16.36 1:47.18 Web Content 3256 cer 20 0 2127024 432164 21292 33792 S 0.000 10.98 1:12.28 Web Content 3381 cer 20 0 1764236 75244 5808 33556 S 0.000 1.913 0:19.27 Web Content 3397 cer 20 0 1688972 28232 10860 45152 S 0.000 0.718 0:07.78 Web Content 2703 root 20 0 449644 20676 7624 15812 S 0.896 0.526 43:45.22 X 4067 cer 20 0 555912 15100 8856 2084 S 0.597 0.384 16:35.73 panel-18-weathe 4038 cer 20 0 1038948 13212 4168 20572 S 0.299 0.336 2:28.46 xfce4-terminal 4003 cer 20 0 869880 10656 876 7360 S 0.000 0.271 0:10.61 xfdesktop 4319 cer 20 0 865912 7872 3344 4908 S 0.000 0.200 1:42.28 contarcorreo 4269 cer 39 19 918436 7584 588 86772 S 0.000 0.193 0:36.22 tracker-miner-f

...

Tasks: 303 total, 2 running, 301 sleeping, 0 stopped, 0 zombie %Cpu0 : 0.2 us, 23.4 sy, 0.0 ni, 0.0 id, 75.0 wa, 0.0 hi, 1.4 si, 0.0 st %Cpu1 : 0.1 us, 35.7 sy, 0.0 ni, 0.0 id, 64.2 wa, 0.0 hi, 0.1 si, 0.0 st

As Berny pointed out yesterday, that system is really busy swapping.

Certainly, but why? I used the laptop, watching TV with it, till some time, then left it alone and went to sleep. In the morning, between 09:30 and 9:49, it crashed.

...

KiB Mem : 3934240 total, 103068 free, 3508588 used, 322584 buff/cache KiB Swap: 6289412 total, 4794296 free, 1495116 used. 21052 avail Mem

So 4Gb main memory and 1.4Gb in swap. That's quite heavy usage, for a desktop - I don't think I have any machine that comes anywhere near that. Just for comparison.

A laptop. It was "running": thunderbird, firefox, chrome. That's all. The rest, desktop compliments.

...

Ok, I wait for suggestions on what to do next.

Don't run so many firefox and chrome processes ? There is a Firefox setting to disable running a process per tab, for instance. WE have had to disable it because of some CSS problem. I'll see if I can find it.

I remember. The problem is not the load. CPU before this was iddling. It is something that is handlinf the load badly. Like requesting out of swap a chunk that when it runs requests some other chunk just now swapped out, alternating. Look at this machine: top - 20:24:03 up 17 days, 9:21, 1 user, load average: 0,72, 0,75, 0,63 Tasks: 575 total, 1 running, 573 sleeping, 0 stopped, 1 zombie %Cpu(s): 6,4 us, 1,9 sy, 0,0 ni, 91,3 id, 0,4 wa, 0,0 hi, 0,0 si, 0,0 st KiB Mem: 8174456 total, 6116724 used, 2057732 free, 36408 buffers KiB Swap: 25165820 total, 4700532 used, 20465288 free. 1437660 cached Mem I'm using it to write this post, it works fine. -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)

On 12/12/2018 09.22, Per Jessen wrote:

Carlos E. R. wrote:

...

On 11/12/2018 15.37, Per Jessen wrote:

...

Carlos E. R. wrote:

...

...

Now I typed a '1' to see each core load. Waiting...

top - 12:18:07 up 4 days, 10:16, 5 users, load average: 41.77, 42.35, 43.84

That's quite a busy desktop you have there.

No, that's while it is crashed. I had been iddling for 12 hours, waiting for it to crash, doing nothing.

So, the machine is not doing anything, then suddenly it goes hyperactive and locks up due to swapping going mad? Something must be triggering that. Maybe that's where you need to start looking?

Well, that's precisely why I ask for help :-) This night it hasn't crashed. What did I do? Remove btrfsmaintenance. The locks seemed to happen after org.opensuse.Snapper activated. But the job is still activated: <3.6> 2018-12-12 10:00:24 minas-tirith dbus-daemon 1360 - - [system] Successfully activated service 'org.opensuse.Snapper' <1.6> 2018-12-12 10:45:01 minas-tirith run-crons 21690 - - leafnode.cron: OK <1.6> 2018-12-12 10:45:01 minas-tirith run-crons 21690 - - mine-clocksync: OK <3.6> 2018-12-12 11:00:24 minas-tirith systemd 1 - - Started Timeline of Snapper Snapshots. <3.6> 2018-12-12 11:00:24 minas-tirith dbus-daemon 1360 - - [system] Activating service name='org.opensuse.Snapper' requested by ':1.459' (uid=0 pid=22050 comm="/usr/lib/snapper/systemd-helper --timeline ") (using servicehelper) <3.6> 2018-12-12 11:00:24 minas-tirith dbus-daemon 1360 - - [system] Successfully activated service 'org.opensuse.Snapper' <3.6> 2018-12-12 11:01:32 minas-tirith smartd 1358 - - Device: /dev/sda [SAT], SMART Usage Attribute: 194 Temperature_Celsius changed from 63 to 62 <1.6> 2018-12-12 11:45:01 minas-tirith run-crons 22930 - - leafnode.cron: OK <1.6> 2018-12-12 11:45:01 minas-tirith run-crons 22930 - - mine-clocksync: OK <3.6> 2018-12-12 12:00:24 minas-tirith systemd 1 - - Started Timeline of Snapper Snapshots. <3.6> 2018-12-12 12:00:24 minas-tirith dbus-daemon 1360 - - [system] Activating service name='org.opensuse.Snapper' requested by ':1.473' (uid=0 pid=23292 comm="/usr/lib/snapper/systemd-helper --timeline ") (using servicehelper) <3.6> 2018-12-12 12:00:24 minas-tirith dbus-daemon 1360 - - [system] Successfully activated service 'org.opensuse.Snapper' So maybe I need to remove something else.

...

...

...

Ok, I wait for suggestions on what to do next.

Don't run so many firefox and chrome processes ? There is a Firefox setting to disable running a process per tab, for instance. WE have had to disable it because of some CSS problem. I'll see if I can find it.

I remember.

The setting is "browser.tabs.remote.autostart.2", which all of our browsers used to have, but deleting it now does not make the CSS problem re-appear. Maybe it got fixed.

I can try to see if that uses less memory. The name of the setting is obscure.

...

The problem is not the load. CPU before this was iddling.

I meant memory load.

Well, Yesterday night you can see one post where I stressed it with several programs: thunderbird, firefox, chrome, kodi... no crash, watching a movie and writing a post, machine very responsive. I have seen it with 2 gigs in swap, double that of the crash, no problem. -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)

On 12/12/2018 13.53, Per Jessen wrote:

Carlos E. R. wrote:

...

This night it hasn't crashed. What did I do? Remove btrfsmaintenance. The locks seemed to happen after org.opensuse.Snapper activated.

But the job is still activated: [snip]

So maybe I need to remove something else.

If you want to get rid of snapper, try removing snapper :-)

Done. minas-tirith:~ # rpm --erase snapper error: Failed dependencies: snapper = 0.5.6 is needed by (installed) snapper-zypp-plugin-0.5.6-lp150.3.6.1.noarch snapper is needed by (installed) yast2-snapper-4.0.4-lp150.2.3.1.x86_64 minas-tirith:~ # rpm --erase snapper snapper-zypp-plugin yast2-snapper Removed /etc/systemd/system/timers.target.wants/snapper-cleanup.timer. Removed /etc/systemd/system/timers.target.wants/snapper-timeline.timer. minas-tirith:~ # -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)

On 12/12/2018 16.10, Liam Proven wrote:

On 12/12/2018 14:17, Carlos E. R. wrote:

...

minas-tirith:~ # rpm --erase snapper error: Failed dependencies: snapper = 0.5.6 is needed by (installed) snapper-zypp-plugin-0.5.6-lp150.3.6.1.noarch snapper is needed by (installed) yast2-snapper-4.0.4-lp150.2.3.1.x86_64 minas-tirith:~ # rpm --erase snapper snapper-zypp-plugin yast2-snapper Removed /etc/systemd/system/timers.target.wants/snapper-cleanup.timer. Removed /etc/systemd/system/timers.target.wants/snapper-timeline.timer. minas-tirith:~ #

Um. Not using Zypper? Doesn't that incur the problem of dependency-satisfaction?

No, it just says that deps are satisfied or not. If not, supply what is missing on the command line. To run zypper on this machine I must mount something on nfs first. :-) -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)

Le 12/12/2018 à 16:30, Ken Schneider - openSUSE a écrit :

The same for disable. snapper doesn't run any longer.

or use yast install, dependencies are easier to set, and you can taboo snapper easily jdd -- http://dodin.org -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 12/12/2018 11.36, Liam Proven wrote:

On 11/12/2018 21:13, Carlos E. R. wrote:

...

Certainly, but why?

Because it only has 4 GB of RAM. That's not really enough for a general-purpose desktop computer these days, IMHO. It's definitely not enough for heavy use of *both* Chrome *and* Firefox. In fact I would go so far as to say it's not enough for heavy use of Chrome on its own.

No, that does not explain it. Yesterday I used it with even more programs and nothing happened, the machine was very responsive. There was 0.9 GB in swap. I have used that machine and others with same RAM with 2.5 GB of swap, no issues. My desktop machine has now 4.3 GB in swap, and it feels happy (although it has 8 GB ram). Remember, the crash happens when I'm sleeping and the machine is doing nothing. Programs remain opened, but not actively doing anything. No, this is a peculiar situation that is not explained that easily.

Stick to Firefox only -- it's a bit more memory-efficient.

I don't like Chrome, I use firefox for my browsing. But I *must* use chrome to watch Amazon Prime Video and cast it to a Chromecast device, that doesn't work on firefox AFAIK. -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)

On 12/12/2018 13.07, Liam Proven wrote:

On 12/12/2018 12:47, Carlos E. R. wrote:

...

No, that does not explain it.

I'm not _trying_ to explain it. I can't.

All I'm saying is that I don't think this will be easily fixed. I am currently going through the pain of moving to a new install of oS on a new PC because it was cheaper than getting a RAM upgrade for my existing one. 8 GB was no longer adequate for my workload and it kept disk-thrashing, and I am not a hugely demanding user.

Linux (the kernel) keeps growing. So does the userland. Especially so do browsers. And memory leaks are everywhere, which is the sort of thing that is fuelling the rise of "devops" and containers: it's virtually impossible to troubleshoot all the issues and leaks in modern, super-complex stacks, so fire them up when needed, and if anything goes wrong, tear them down, destroy them and fire up new instances. It's easier than trying to get them to be stable and just work.

This does not help us desktop users, but sadly, desktop Linux is a tiny minority niche and will never be anything else. :-(

:-(

...

I don't like Chrome, I use firefox for my browsing. But I *must* use chrome to watch Amazon Prime Video and cast it to a Chromecast device, that doesn't work on firefox AFAIK.

So, if you only have 4 GB RAM, when you want to screencast something with Chrome, shut down Firefox.

But that's not it. See: I work for some time with the machine, doing things happily with that workload. I leave it alone for an hour or a night. The screen saver kicks in (black). I go back, touch a key, no response. What happened? How can the computer be trashing with the same programs that were working an hour before, but when I stop using it, it crashes, the program or the kernel goes mad? I have just posted some calculations, a comparison of the memory load when crashed and when normal. Same programs loaded.

Or get more RAM. :-(

it is a laptop, that's impossible, AFAIK :-( What I did was get faster swap. SSD. -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)

On 12/12/2018 13.45, jdd@dodin.org wrote:

Le 12/12/2018 à 13:29, Carlos E. R. a écrit :

...

I leave it alone for an hour or a night. The screen saver kicks in (black). I go back, touch a key, no response.

What happened?

*the screen saver* ?? did it try to hybernate?

No. At least, it shouldn't. just power off the display. Hibernation is manual, or when battery goes low. Some of the times it happened, I was sitting there, but watching TV, not using the computer. But those times I had to reboot to recover, now I know that I can ssh and find out things. -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)

On 12/12/2018 13.51, Liam Proven wrote:

On 12/12/2018 13:29, Carlos E. R. wrote:

...

...

This does not help us desktop users, but sadly, desktop Linux is a tiny minority niche and will never be anything else. :-(

:-(

Well, yes, agreed.

But this is how and why Apple and Microsoft make money. They put a huge amount of R&D effort into desktop OSes.

The Linux companies don't, because Linux desktops don't. Servers do, so we have some not-very-well tested, barely-integrated desktops on top of a server OS.

Ubuntu used to put more effort into desktop integration than almost anyone -- SUSE is a server vendor, RH is a server vendor, and there basically isn't anyone else.

All the pure-play desktop Linux vendors (Corel, Caldera, Xandros, Lindows/Linspire, Mandriva) are dead.

Sadly, there's a reason.

ChromeOS is about it, and part of the reason it works is that it's so massively cut-down compared to a normal desktop OS.

...

But that's not it.

See: I work for some time with the machine, doing things happily with that workload.

I leave it alone for an hour or a night. The screen saver kicks in (black). I go back, touch a key, no response.

What happened?

It's thrashing. I think we've established that.

That's the symptom, not the cause.

...

How can the computer be trashing with the same programs that were working an hour before, but when I stop using it, it crashes, the program or the kernel goes mad?

Probably because something is leaking memory.

The numbers do not add up.

...

...

Or get more RAM. :-(

it is a laptop, that's impossible, AFAIK :-(

What I did was get faster swap. SSD.

So get a newer laptop.

I did! It also has 4 GB of RAM. It is smaller and lighter, I can travel with it, not heavy. And it does not exhibit these symptoms (so far: I'll test for it. But I have been using it a lot recently). I simply did not find something else at a reasonable price and small footprint. Yet at the time I did not know about this problem.

I had a lovely big Toshiba Satellite Pro P300A, but it has an old chipset so it takes DDR2 and while upgrading it from 3 GB to 4GB was about £5, going from 4 GB to 8 GB would be about £80.

I will have to study it. It is a "compaq presario CQ61-330SS". https://www.kingstonmemoryshop.co.uk/laptop/hp-compaq/presario-cq-series/hp-... Says it can be maxed to 8 GiB. I must consider it, then. I think I would need two 4 GiB modules. I think the price is 79.99£ each. Hum...

So earlier this year, I sold it to a friend, for about £100. I kept its SSD.

And I replaced it with the Thinkpad X220 I bought last year for £150.

The Satellite was my testbed machine, after I replaced it with a 2nd-hand Mac mini.

Now, the X200 that the X220 replaced is my testbed box. (Currently it runs IBM PC DOS 7.1, Bluebottle/A2 and Haiku, with Devuan for experimenting with VMs etc. I have failed to get eComStation to install.)

Second-hand laptops are pretty cheap. Thinkpads are tough, so 2nd hand Thinkpads are a good deal. They are also well-tested with Linux, and everything usually Just Works™.

I don't know how to buy second hand laptops here (Spain, not the capital city). There was a shop, it disappeared.

I got mine from Morgan's. They shipped to Czechia.

https://www.morgancomputers.co.uk/c/599/Laptop-Sale/

Remote purchase? Hum. I don't like that. But I'll take a note.

Tier 1 are also good.

https://www.tier1online.com/

I am sure there are similar vendors in Spain and elsewhere in Europe. Latin America, I don't know -- but there's eBay and so on.

I don't like eBay... -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)

Le 12/12/2018 à 16:15, Liam Proven a écrit :

On 12/12/2018 14:12, Carlos E. R. wrote:

...

I don't know how to buy second hand laptops here (Spain, not the capital city). There was a shop, it disappeared.

https://www.vibbo.com/ search on your vicinity so you buy what you see jdd -- http://dodin.org -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 12/12/2018 17:51, jdd@dodin.org wrote:

a huge amount of efforts *selling* OS :-(

Actually, no. Mac OS X -- or "macOS" as the company prefers now -- is a free download. Of course you are only _supposed_ to run it on Apple hardware, but I have built a Hackintosh and I used it for some years as my main computer. It worked very well indeed. Windows 10 is not officially free, but it was a free _upgrade_ and it's close. The free upgrades from Windows 7/8/8.1 still work, and you can download an ISO from microsoft.com and run it unregistered and it works well -- it is less crippled than older versions were if run unregistered, because MS worked out that crippled illegal versions were a reservoir and vector for malware. Apple's business is selling _hardware_. Microsoft's business is selling Office and server software. The end-user OSes are all but freeware. (Yes, I *know* freeware != FOSS.)

for example selling to French Defense Minister a contract for more than 188500 computers, *hardware* an software. No linux company can compete :-(

Sadly, there are solid commercial/technical reasons that things like this happen: https://www.theregister.co.uk/2018/01/04/munich_linux_costs_ownership/ I find it very interesting that, for instance, ChromeOS machines are selling into a few enterprises that use the commercial gMail/gApps offering. All the companies offering desktop Linuxes spent a huge amount of time and effort on trying to look like, work like, and be compatible with, Windows. Rich email clients, rich productivity suites, over-engineered feature-heavy chat systems. Why, because everyone knows how to use Windows, and Office, and Outlook+Exchange is *the* groupware solution. Google looks at this and goes "nope, not going to try." It does a good-enough, not-very-customisable web groupware system. It makes sure it has good 3rd party support. It makes a good-enough, not-very-rich web app system. They're web apps, so they *can't* compete with the best rich local apps. Instead, they play to their strengths: collaborative editing. And it offers a sort of web-aware thin client, based on FOSS, that uses Google's single-sign-on, and does basically nothing locally. Real engineering effort is put into just one part: a very smart, low-tech but robust, self-updating system. Package management, desktops, all the stuff that distro vendors do, is basically ignored. And it sells hundreds of millions of the resulting devices. I think it was a very smart move. I remain surprised that basically nobody in the FOSS world has tried to copy it. -- Liam Proven - Technical Writer, SUSE Linux s.r.o. Corso II, Křižíkova 148/34, 186-00 Praha 8 - Karlín, Czechia Email: lproven@suse.com - Office telephone: +420 284 241 084 -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 13/12/2018 13:16, jdd@dodin.org wrote:

...

Windows still sell it's OS in new computers (nearly forced selling)

Well, yes, it does, true. And so does Apple. But, for instance, you can build your own desktop computer and run either OS on it. Me, I like the simplicity of macOS, so although I have the skills to build my own PC, and have built many PCs, I just bought a used Mac. This year, I replaced it with a newer used Mac. It's pretty, quiet, and it just works. I don't like Apple's modern keyboards and pointing devices, though, so I use a 1980s mechanical Apple keyboard (through an interface convertor). My laptops are cheap used Thinkpads. It saves me a lot of money and I am happy with the results.

and you are true, but no FOSS company have the Google power:-(

It's not about Google's power. Anyone could have done a 3rd party FOSS implementation of Microsoft's groupware, but nobody did. Why not? Well, I think it's because Unix people tend to like simple old-fashioned tools like plain-text email over IMAP, IRC and so on. Anyone could have done a network directory and admin tool like Active Directory. Why didn't they? Because Unix people love their text config files and have strong preferences about things like packaging systems. Result, now, we have systemd and lots of people hate it, and we _still_ don't have a modern working network directory and admin tool for Unix. Anyone could have done standards-based single-sign-on, and integrated it with the above two things. But nobody did, because for Linux vendors, the money is in cloud servers, not stuff like that. Result, Microsoft remained in control of its market segment, Apple of its, and Google has done a sort of end-run around them both which might ultimately win out and outsell both of the big 2 proprietary vendors. And the most ironic thing of all, Google's offerings are based on Linux after all! -- Liam Proven - Technical Writer, SUSE Linux s.r.o. Corso II, Křižíkova 148/34, 186-00 Praha 8 - Karlín, Czechia Email: lproven@suse.com - Office telephone: +420 284 241 084 -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 12/12/2018 14.37, Anton Aylward wrote:

On 2018-12-12 7:29 a.m., Carlos E. R. wrote:

...

I leave it alone for an hour or a night. The screen saver kicks in (black). I go back, touch a key, no response.

Googling I find that Xscreensaver has repeatedly reported memory leaks over the years.

It doesn't show in the top list of memory users when I run top. Maybe because I touched the keyboard? No, if no password in some seconds it goes back to dark. Dunno. -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)

On 12/12/2018 12.21, Anton Aylward wrote:

On 2018-12-11 3:13 p.m., Carlos E. R. wrote:

...

I'm using it to write this post, it works fine.

Just to erasure me .. What kernels are running on each of these machines?

Defaults. minas-tirith:~ # uname -a Linux minas-tirith.valinor 4.12.14-lp150.12.25-default #1 SMP Thu Nov 1 06:14:23 UTC 2018 (3fcf457) x86_64 x86_64 x86_64 GNU/Linux minas-tirith:~ # That's the problem machine, a laptop. -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)

On 11/12/2018 18.14, Andrei Borzenkov wrote:

11.12.2018 14:37, Carlos E. R. пишет:

...

Ah, got /proc/meminfo in another way:

minas-tirith:~ # cat /proc/meminfo > meminfo minas-tirith:~ # cat meminfo MemTotal: 3934240 kB MemFree: 106524 kB MemAvailable: 24012 kB

Your system has no free memory to do anything useful. So any attempt to start any new program will result in attempt to free something, then system must recall pages it just freed from disk again and to do it it needs to to find free memory again ... you get an idea.

Yes, I guessed that, but I don't know how it gets there. Procedure: watch TV on laptop. Stop TV. Go to sleep. Come back, locked. I had running: thunderbird (crashed), firefox, chrome. Thunderbird had been told to exit, days ago; but something remained loaded refusing to exit:

PID USER PR NI VIRT RES SHR SWAP S %CPU %MEM TIME+ COMMAND 4027 cer 20 0 1943072 3136 792 33888 S 0.043 0.080 0:20.19 thunderbird-bin

...

Buffers: 1596 kB Cached: 278208 kB SwapCached: 97988 kB Active: 2949780 kB Inactive: 618720 kB Active(anon): 2945904 kB Inactive(anon): 589672 kB Active(file): 3876 kB Inactive(file): 29048 kB

Most memory is consumed by some application(s) allocating (and actually using) large amount of memory. It is simply impossible that all loaded application binaries amount to just 33MB. On my system with rather static load active+inactive file is 4.8GB, and I have just Chromium + Thunderbird + Deluge + several terminal windows as part of user session.

Look, list of top processes sorted by memory (RES) just after killing Thunderbird and Chrome, so with responsive system: top - 13:31:23 up 4 days, 11:29, 5 users, load average: 0.85, 19.32, 34.43 Tasks: 271 total, 1 running, 270 sleeping, 0 stopped, 0 zombie %Cpu0 : 1.5 us, 1.3 sy, 0.0 ni, 97.1 id, 0.0 wa, 0.0 hi, 0.1 si, 0.0 st %Cpu1 : 1.5 us, 0.7 sy, 0.0 ni, 97.7 id, 0.0 wa, 0.0 hi, 0.1 si, 0.0 st KiB Mem : 3934240 total, 1306420 free, 2123600 used, 504220 buff/cache KiB Swap: 6289412 total, 5290168 free, 999244 used. 1449112 avail Mem PID USER PR NI VIRT RES SHR SWAP S %CPU %MEM TIME+ COMMAND 3256 cer 20 0 2543792 805272 15840 30120 S 0.000 20.47 0:53.40 Web Content 3342 cer 20 0 2336788 640792 37524 32844 S 0.000 16.29 1:24.18 Web Content 3164 cer 20 0 9487996 350420 61860 52756 S 0.439 8.907 7:32.39 firefox 3381 cer 20 0 1764236 77596 6964 32360 S 0.000 1.972 0:19.15 Web Content 2703 root 20 0 449308 24744 11532 15652 S 1.023 0.629 40:07.73 X 3397 cer 20 0 1688972 20756 2840 44604 S 0.000 0.528 0:07.70 Web Content 4038 cer 20 0 1038948 15516 4120 18216 S 0.146 0.394 2:02.10 xfce4-terminal 4067 cer 20 0 555352 12760 6484 2044 S 0.439 0.324 14:42.96 panel-18-weathe 4003 cer 20 0 869880 11492 1344 6992 S 0.000 0.292 0:10.20 xfdesktop 4319 cer 20 0 865652 7916 3404 4680 S 0.000 0.201 1:33.80 contarcorreo 4269 cer 39 19 918436 7792 780 86756 S 0.000 0.198 0:34.50 tracker-miner-f 4030 cer 20 0 875056 7192 2900 4028 S 0.585 0.183 18:49.47 gkrellm 22832 root 20 0 18420 6700 2188 508 S 0.000 0.170 0:00.35 bash 3989 cer 20 0 747576 6016 3700 5944 S 0.000 0.153 0:16.09 xfce4-panel 3981 cer 20 0 1003288 5048 112 6780 S 0.000 0.128 0:07.42 Thunar Nothing is really using a lot memory, unless we count virtual memory. What is above sums 2 GiB. Now, virt is 25GiB. Look, I had killed Thunderbird and Chrome, which were: PID USER PR NI VIRT RES SHR SWAP S %CPU %MEM TIME+ COMMAND 4027 cer 20 0 1943072 3136 792 33888 S 0.043 0.080 0:20.19 thunderbird-bin 12526 cer 20 0 1453444 137488 2816 14312 D 3.001 3.495 145:56.66 chrome 12670 cer 20 0 1703752 151676 5356 15124 D 0.749 3.855 104:20.03 chrome 12563 cer 20 0 2089092 63880 4524 243644 D 0.400 1.624 93:40.60 chrome 15952 cer 20 0 876624 118512 16556 21176 D 0.289 3.012 6:02.88 chrome 14060 cer 20 0 538600 4968 176 9852 D 0.238 0.126 140:14.77 chrome I should have queried sort by RES before starting killing, but you can guess that each operation took minutes. But I have it on the first incident on this thread: top - 21:48:19 up 1 day, 19:46, 5 users, load average: 9.18, 11.24, 14.59 Tasks: 291 total, 4 running, 287 sleeping, 0 stopped, 0 zombie %Cpu(s): 2.4 us, 18.4 sy, 0.0 ni, 9.9 id, 68.1 wa, 0.0 hi, 1.1 si, 0.0 st KiB Mem : 3934240 total, 106132 free, 3417824 used, 410284 buff/cache KiB Swap: 6289412 total, 4956028 free, 1333384 used. 56300 avail Mem PID USER PR NI VIRT RES SHR SWAP USED S %CPU %MEM TIME+ COMMAND 4355 cer 20 0 9.857g 1.168g 33104 0 1.168g D 0.946 31.13 5:11.20 firefox 4603 cer 20 0 2189304 497600 20004 0 497600 S 0.000 12.65 0:20.22 Web Content 4476 cer 20 0 2121180 369828 4 0 369828 S 0.000 9.400 0:31.32 Web Content 12670 cer 20 0 1645688 204172 16424 24272 228444 S 2.208 5.190 60:26.25 chrome 4574 cer 20 0 1800896 123752 4776 0 123752 S 0.000 3.146 0:36.18 Web Content 4616 cer 20 0 1748260 113256 4444 0 113256 S 0.000 2.879 0:09.17 Web Content 12526 cer 20 0 1436224 93956 2964 20672 114628 D 4.416 2.388 84:47.28 chrome 15650 cer 20 0 896392 84652 12128 31548 116200 D 1.262 2.152 10:37.98 chrome 16330 cer 20 0 836828 81516 19036 29160 110676 D 0.631 2.072 3:20.19 chrome 12563 cer 20 0 2077116 79692 6032 228036 307728 S 0.000 2.026 61:25.77 chrome 15996 cer 20 0 827280 77852 15320 24792 102644 S 0.631 1.979 3:16.00 chrome 15952 cer 20 0 832136 77220 2944 26892 104112 S 0.000 1.963 3:07.12 chrome 15265 cer 20 0 845048 57020 728 33540 90560 S 0.000 1.449 11:23.63 chrome 16400 cer 20 0 793004 41616 744 26940 68556 S 0.315 1.058 1:22.59 chrome 2703 root 20 0 450280 41456 28532 15100 56556 D 0.631 1.054 22:37.38 X 4003 cer 20 0 869880 10892 8 6192 17084 S 0.000 0.277 0:05.23 xfdesktop 4038 cer 20 0 1036900 10676 28 17256 27932 S 0.000 0.271 0:37.96 xfce4-terminal 4030 cer 20 0 875056 8264 1888 1888 10152 S 0.631 0.210 9:59.47 gkrellm firefox was using 1.168g, but that is not "a lot", I have seen it at two and the machine huuging away happily and responsive.

Again - it means that any attempt to do anything will require loading binaries from disk. This alone would account for high wait time observed earlier.

What is your swappiness settings?

Defaults. Never touched it. minas-tirith:~ # cat /proc/sys/vm/swappiness /proc/sys/vm/vfs_cache_pressure /proc/sys/vm/min_free_kbytes /proc/sys/vm/watermark_scale_factor 60 100 67584 10 minas-tirith:~ # /etc/sysctl.conf: ### converted from /etc/sysconfig/sysctl at Mon, 06 Jan 2014 02:32:57 +0100 net.ipv4.tcp_syncookies = 1 net.ipv4.ip_forward = 0 net.ipv6.conf.all.forwarding = 0 net.ipv4.tcp_ecn = 0 # net.ipv6.conf.all.disable_ipv6 = 1

...

Unevictable: 80 kB Mlocked: 80 kB SwapTotal: 6289412 kB SwapFree: 4813752 kB Dirty: 104 kB Writeback: 0 kB AnonPages: 3282560 kB Mapped: 88652 kB Shmem: 246612 kB Slab: 108992 kB SReclaimable: 42936 kB SUnreclaim: 66056 kB KernelStack: 13696 kB PageTables: 79240 kB NFS_Unstable: 0 kB Bounce: 0 kB WritebackTmp: 0 kB CommitLimit: 8256532 kB Committed_AS: 13514900 kB VmallocTotal: 34359738367 kB VmallocUsed: 0 kB VmallocChunk: 0 kB HardwareCorrupted: 0 kB AnonHugePages: 399360 kB ShmemHugePages: 0 kB ShmemPmdMapped: 0 kB HugePages_Total: 0 HugePages_Free: 0 HugePages_Rsvd: 0 HugePages_Surp: 0 Hugepagesize: 2048 kB DirectMap4k: 644416 kB DirectMap2M: 3450880 kB minas-tirith:~ #

Ok, I wait for suggestions on what to do next.

Your system is overloaded. Either do not run programs that consume such amount of RAM or add RAM. Tweaking swappiness may help to prevent pushing programs out of memory.

No, that's not it. I was using happily that machine during the night, then I left it alone, not closing any program, I went to sleep, and some 12 hours it crashed. Only cron and timers were changing things. I have been using the same load a long time, with 42.3, before upgrading to 15.0 I'll try removing "btrfsmaintenance". -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)

On 12/12/2018 02.46, Carlos E. R. wrote:

On 11/12/2018 21.13, Carlos E. R. wrote:

...

On 11/12/2018 18.14, Andrei Borzenkov wrote:

...

11.12.2018 14:37, Carlos E. R. пишет:

...

...

...

Your system is overloaded. Either do not run programs that consume such amount of RAM or add RAM. Tweaking swappiness may help to prevent pushing programs out of memory.

No, that's not it. I was using happily that machine during the night, then I left it alone, not closing any program, I went to sleep, and some 12 hours it crashed.

Look, this minute I have active:

Firefox (same pages as yesterday, no change). Chrome (Amazon Prime Video page). Thunderbird, actively writing this email. Kodi, playing a video.

The system is perfectly responsive, with a much higher load than yesterday. Observe this:

top - 03:06:08 up 5 days, 1:04, 5 users, load average: 0.27, 0.32, 0.42 Tasks: 294 total, 2 running, 292 sleeping, 0 stopped, 0 zombie %Cpu(s): 6.6 us, 3.6 sy, 0.0 ni, 89.5 id, 0.0 wa, 0.0 hi, 0.3 si, 0.0 st KiB Mem : 3934240 total, 819244 free, 1883652 used, 1231344 buff/cache KiB Swap: 6289412 total, 5317560 free, 971852 used. 1521480 avail Mem PID USER PR NI VIRT RES SHR SWAP USED S %CPU %MEM TIME+ COMMAN ... 4183 cer 39 19 0.250t 13836 7472 2448 16284 S 0.000 0.352 0:08.65 baloo_file This single process says it is using 1/4 of a terabyte virtual memory! I know this is not real memory, but still... -- Cheers / Saludos, Carlos E. R. (from oS Leap 15.0 x86_64 (Minas Tirith))

On 2018-12-11 9:08 p.m., Carlos E. R. wrote:

No, that's not it. I was using happily that machine during the night, then I left it alone, not closing any program, I went to sleep, and some 12 hours it crashed.

That sounds to me like a memory leak. In fact there's a lot in this thread that reeks of a memory leak. Further, I'm getting a system freezing and an insanely high load factors. I hotkey to a VT where I've logged in and run 'w' which is comparatively lightweight and see load factors of between 9 and 33.. Killing firefox or dolphin or thunderbird relieves it. Thee dolphin had a lot of tabs and takes memory for rendering; I'm not sure about the libraries it uses. TB and FF are GTK based and both also use HTML rendering libraries. Perhaps it is a HTML page or image or something... I've massively reduced the number of FF tabs and things seem to be holding. So I'm suspicious that there is a memory leak in the HTML rendering software somewhere along the line. -- A: Yes. > Q: Are you sure? >> A: Because it reverses the logical flow of conversation. >>> Q: Why is top posting frowned upon? -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 2018-12-12 7:21 a.m., Carlos E. R. wrote:

top - 21:48:19 up 1 day, 19:46, 5 users, load average: 9.18, 11.24, 14.59

Yikes, look at those load averages. WHY? You've got plenty of free swap even if something is using memory. What are things waiting on There's another tool to look at the CPU/wait queues, but I forget what it is.

Tasks: 291 total, 4 running, 287 sleeping, 0 stopped, 0 zombie %Cpu(s): 2.4 us, 18.4 sy, 0.0 ni, 9.9 id, 68.1 wa, 0.0 hi, 1.1 si, 0.0 st KiB Mem : 3934240 total, 106132 free, 3417824 used, 410284 buff/cache KiB Swap: 6289412 total, 4956028 free, 1333384 used. 56300 avail Mem

-- A: Yes. > Q: Are you sure? >> A: Because it reverses the logical flow of conversation. >>> Q: Why is top posting frowned upon? -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 2018-12-12 1:56 p.m., Carlos E. R. wrote:

...

What are things waiting on On memory to be freed by swaping out, but what was swapped out was also needed, so undo, in a death cycle.

...

There's another tool to look at the CPU/wait queues, but I forget what it is. htop?

No. htop _can_ tell you what queue, but there is a text-mode tool that lists all the wait and everything waiting on them. Option for output on XML or some sort of array that other programs can digest. I forget what it is. complicated set op parameters :-( https://cs.unc.edu/~blate/courses/comp530H_F14/pdf/Sync_Schedulie.pdf -- A: Yes. > Q: Are you sure? >> A: Because it reverses the logical flow of conversation. >>> Q: Why is top posting frowned upon? -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 2018-12-11 9:08 p.m., Carlos E. R. wrote:

This single process says it is using 1/4 of a terabyte virtual memory! I know this is not real memory, but still...

Indeed, and that too can be indicative of a memory leak. Yes, what matters is the working set, but even that may result in paging. I admit there are some applications that might want to use that 1/4 of a terabyte virtual memory, some photo editing for example, but common use, email, web browsing, regular editing shouldn't. Its not as if you are composing a coffee table book with double A3 page spreads. Or are you? In this instance, the 'baloo_file' there isn't much CPU activity. It may be that it is running in some CPU-limited framework. If something like this really freaks you out then there us always the possibility of starting it under a ULIMIT-ed shell (see man bashbuiltins) So go back to the 'top' command and see what is /really/ consuming resources, what is thrashing, what is leaking memory. -- A: Yes. > Q: Are you sure? >> A: Because it reverses the logical flow of conversation. >>> Q: Why is top posting frowned upon? -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 12/12/2018 06:36 AM, Carlos E. R. wrote:

The only thing I did was uninstall btrfsmaintenance.

And we were told it did nothing if you didn't have a butter install.... -- David C. Rankin, J.D.,P.E.

On 13/12/2018 13.11, Liam Proven wrote:

On 13/12/2018 09:05, David C. Rankin wrote:

...

And we were told it did nothing if you didn't have a butter install....

It doesn't. But AFAICT, it still starts, looks around, decides it has nothing to do, and quits.

And that takes time.

Have a look at this one: http://bugzilla.opensuse.org/show_bug.cgi?id=1096589 -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)

On 12/12/2018 13.36, Carlos E. R. wrote:

On 12/12/2018 12.17, Anton Aylward wrote:

...

On 2018-12-11 9:08 p.m., Carlos E. R. wrote:

...

...

So go back to the 'top' command and see what is /really/ consuming resources, what is thrashing, what is leaking memory.

Well, as I said, I have one terminal running "top" like this:

top -o RES -b -n 4000 -d 300 > somefile

to log the crash. But tonight, it did not crash. It is working just fine, with the same load that crashed two days in a row.

The only thing I did was uninstall btrfsmaintenance.

Well, no more crashes so far. Go figure. I will watch it, though. Since yesterday night, the machine is running Thunderbird, Firefox, Chrome, and Kodi, no issues. Perfectly responsive with that large load. Less than 3 GB of ram are used by processes, leaving half a gig free, another half for buffers/cache, and half available. This is the normal situation under "load". More than 1 gig of swap used. (sorting by RES) top - 15:00:31 up 22:56, 5 users, load average: 0.24, 0.28, 0.21 Tasks: 292 total, 2 running, 290 sleeping, 0 stopped, 0 zombie %Cpu(s): 5.6 us, 1.6 sy, 0.3 ni, 92.4 id, 0.0 wa, 0.0 hi, 0.1 si, 0.0 st KiB Mem : 3934076 total, 497804 free, 2865376 used, 570896 buff/cache KiB Swap: 6289412 total, 5016324 free, 1273088 used. 662160 avail Mem PID USER PR NI VIRT RES SHR SWAP USED S %CPU %MEM TIME+ COMMAND 4823 cer 20 0 3593232 715736 33792 327808 0.995g S 1.036 18.19 9:21.17 thunderbird-bin 18262 cer 20 0 9443652 591152 48364 34720 625872 S 2.246 15.03 20:52.13 firefox 18381 cer 20 0 1863888 212596 31224 11120 223716 S 2.783 5.404 16:12.69 Web Content 18440 cer 20 0 1866280 209640 29976 18148 227788 R 2.816 5.329 16:12.65 Web Content 18426 cer 20 0 2130288 186116 21692 21352 207468 S 0.303 4.731 2:17.77 Web Content 5767 cer 20 0 1700176 167088 19996 22292 189380 S 0.283 4.247 44:00.77 chrome 5622 cer 20 0 1227188 103372 19484 22804 126176 S 0.220 2.628 43:34.19 chrome 18503 cer 20 0 1657076 55080 16756 33164 88244 S 0.047 1.400 0:18.83 Web Content 2725 root 20 0 403324 52420 23236 18820 71240 S 1.700 1.332 30:13.84 X 4860 cer 20 0 1039936 28788 3272 5840 34628 S 0.127 0.732 1:08.97 xfce4-terminal 5658 cer 20 0 808596 28588 9356 60956 89544 S 0.007 0.727 50:15.21 chrome 5077 cer 39 19 704984 24788 1144 69180 93968 S 0.007 0.630 0:15.09 tracker-miner-f 2608 news 20 0 33544 17732 2044 0 17732 S 0.030 0.451 0:00.09 leafnode 21196 cer 20 0 664884 13828 4736 4852 18680 S 0.000 0.351 0:00.17 chrome 5134 cer 20 0 953708 13632 3076 10272 23904 S 0.000 0.347 0:02.56 gnote 4874 cer 20 0 481968 13100 6584 1728 14828 S 0.493 0.333 6:45.89 panel-18-weathe 2609 news 20 0 28268 12332 1960 0 12332 S 0.020 0.313 0:00.06 leafnode 4850 cer 20 0 968268 11492 1256 23916 35408 S 0.000 0.292 0:04.01 keepassxc 4992 cer 39 19 0.250t 11360 8012 2324 13684 S 0.003 0.289 0:02.79 baloo_file 4808 cer 20 0 385664 10952 4464 1424 12376 S 0.003 0.278 0:04.85 xfce4-panel 5796 cer 20 0 531956 10832 2656 6020 16852 S 0.000 0.275 63:04.81 chrome -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)

On 17/12/2018 07.42, ken wrote:

On 12/16/18 9:09 AM, Carlos E. R. wrote:

...

Well, no more crashes so far. Go figure. I will watch it, though.

Since yesterday night, the machine is running Thunderbird, Firefox, Chrome, and Kodi, no issues. Perfectly responsive with that large load.

Sorry for being late to the thread.  I've had quite similar problems several times in the past, found them to be due to specific web pages, probably running bad javascript.  When I close the offending webpage, the problem goes away.

For a long time I've been using a Firefox add-in called NoScript.  It allows me to turn on/off specific URLs from which scripts are delivered.  It's a nice app, but would be much better if it had some intelligence to detect when a script was devouring RAM and at least flash a notice of the (potential) problem.  As it is, that task is left to the user... and it's no trivial exercise.  The best that "top" provides is one of the "Web Content" processes, but not which window or tab or site.  The PID numbering might, in more orderly scenarios, offer a plausible guess as to the offending process.  But I haven't yet tried simply killing it... don't know what further hell that might unleash upon an important session.  Javascript coders should test their stuff better before they loose it on the public, as should web developers.  As it is, we all become the beta testers.

That's a possibility, and in my tests I had Firefox opened at the same pages. The suspicious one belongs to my own router. But you see, the laptop can be fine for an hour or ten hours, and then suddenly it goes caput. The issue is not high memory load, but that a lot of that memory tries to be active instead of in swap. -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)

On 17/12/2018 15:19, Anton Aylward wrote:

If I knew the pages, and hence the web site, I could (a) not go there and (b) point the problem out to the webmaster.

This gave me an idea. It might be a silly idea, I don't know. Chrome introduced an innovation years ago that I loved: it puts a little loudspeaker icon on any tab that's playing music. Firefox later copied this. That allows nasty bad-mannered websites with autoplay videos -- such as C|Net -- to be identified and silenced. How about an optional addon that embeds a CPU use histogram into the tab headers, so you can see which tabs are the most CPU-heavy? It's probably not something you'd want on all the time, but for troubleshooting it could be very useful. -- Liam Proven - Technical Writer, SUSE Linux s.r.o. Corso II, Křižíkova 148/34, 186-00 Praha 8 - Karlín, Czechia Email: lproven@suse.com - Office telephone: +420 284 241 084 -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 18/12/2018 13.36, Anton Aylward wrote: Notice that this is tangent on my CPU load problem. I don't mind, just that if you try to fit my symptoms they will not match.

On 2018-12-18 7:07 a.m., Liam Proven wrote:

...

On 17/12/2018 15:19, Anton Aylward wrote:

...

If I knew the pages, and hence the web site, I could (a) not go there and (b) point the problem out to the webmaster.

This gave me an idea. It might be a silly idea, I don't know.

Chrome introduced an innovation years ago that I loved: it puts a little loudspeaker icon on any tab that's playing music. Firefox later copied this.

That allows nasty bad-mannered websites with autoplay videos -- such as C|Net -- to be identified and silenced.

Is that auto-play that requires the net connection or just animated GIF? Some email I get has short animated GIFs; easy to tell 'cos the loop time is very short. But I can imagine longer ones emulating a video that make massive demands of memory and (CPU?) one the way down have hit network bandwidth as much as if they were a video.

The current feature the browser has is simply knowing which tab is producing sound. Just that. When you start firefox and it loads the previous state with several tabs and windows, it is possible that firefox or chrome starts blasting sound at full volume, because one of the one hundred tabs is on a page that automatically loads a video (mind, video, not gif) and it plays automatically with full sound. Or a page that has "background" music. Nasty. It is not a CPU problem, just a sound problem. Good luck finding the nasty tab. Now, an idea based on that one comes below:

...

How about an optional addon that embeds a CPU use histogram into the tab headers, so you can see which tabs are the most CPU-heavy?

Yes, but is that relevant? As far as I can tell the incredibly high load factors are an emergent property of wait times rather than CPU load.

And this is where you are trying to fit my symptoms into this tangent and failing :-)

...

It's probably not something you'd want on all the time, but for troubleshooting it could be very useful.

Shooting some kinds of trouble ...

I would like that addon. Or, even better, open another tab that lists all tabs with a small photo (so that I can click and jump to any of them), and also displays the resources used by each tab (including CPU load) and allows to pause or stop (kill) a particular tab. Freeze a tab would be interesting. But I doubt that FF can say the CPU load of a tab, because it does not assign a process per tab. -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)

On 18/12/2018 14:57, Anton Aylward wrote:

(I'm still on 52.8 for a variety of reasons, most to do with the add-ons)

Try Waterfox. https://www.waterfoxproject.org/en-US/ The security updates of Quantum, but classic add-ons still work. No RPM for openSUSE but if you unpack the tarball into /opt it runs fine. -- Liam Proven - Technical Writer, SUSE Linux s.r.o. Corso II, Křižíkova 148/34, 186-00 Praha 8 - Karlín, Czechia Email: lproven@suse.com - Office telephone: +420 284 241 084 -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 18/12/2018 14:31, Carlos E. R. wrote:

I would like that addon.

:-)

Or, even better, open another tab that lists all tabs with a small photo (so that I can click and jump to any of them),

I believe Opera can do that. I would also suggest that you investigate a vertical tab bar. I find them indispensable. I don't use mainline Firefox much any more but there are some for it now. TressStyleTab works. You have to manually hack your $profile/chrome/userChrome.css file to hide the horizontal tabs and the sidebar header, but then it works. (The hackery is why I favour Waterfox these days. The compiled binary they offer for download runs a treat on openSUSE with no dependencies.)

and also displays the resources used by each tab (including CPU load) and allows to pause or stop (kill) a particular tab.

Yup.

Freeze a tab would be interesting.

Yup.

But I doubt that FF can say the CPU load of a tab, because it does not assign a process per tab.

Ah, good point. -- Liam Proven - Technical Writer, SUSE Linux s.r.o. Corso II, Křižíkova 148/34, 186-00 Praha 8 - Karlín, Czechia Email: lproven@suse.com - Office telephone: +420 284 241 084 -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On Tue, 18 Dec 2018 15:26:31 +0100 Liam Proven wrote:

I opened a few tabs with a middle-click, almost at random.

#3 was this:

https://www.cnet.com/news/scientists-discover-rare-fossil-from-the-big-bang/

This has an autoplay video, unrelated to the story, at bottom right. Currently it is trying to tell me about "the top 5 best tech toys".

Hmm, there's no autoplay video on my machine, because cnet isn't allowed to run javascript. And if I do allow it for some reason, there's still nothing. If I further allow cbsistatic.com then there's a whole load of crap loads, but still no autoplay video. In fact no matter what I do there's no autoplay video, simply because I disabled it globally in Firefox's preferences. :) -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On Tue, 18 Dec 2018 20:34:10 +0100 "Carlos E. R." wrote:

On 18/12/2018 17.20, Dave Howorth wrote:

...

On Tue, 18 Dec 2018 15:26:31 +0100 Liam Proven wrote:

...

I opened a few tabs with a middle-click, almost at random.

#3 was this:

https://www.cnet.com/news/scientists-discover-rare-fossil-from-the-big-bang/

This has an autoplay video, unrelated to the story, at bottom right. Currently it is trying to tell me about "the top 5 best tech toys".

Hmm, there's no autoplay video on my machine, because cnet isn't allowed to run javascript. And if I do allow it for some reason, there's still nothing. If I further allow cbsistatic.com then there's a whole load of crap loads, but still no autoplay video. In fact no matter what I do there's no autoplay video, simply because I disabled it globally in Firefox's preferences. :)

I have disabled autoplay video, yet they play.

I've yet to encounter any example at all that does that. Could you provide a link to one? (Oh, you are using the opensuse packaged 60.3.0esr? Things are different in later versions as a quick search will explain)

The new fashion on some news sites when you open certain pages with videos is that the video is there, loading but stopped (thus using bandwith). As I read the text and page down, a new small window pops up at the bottom left of the page, with the video, sometimes running, and impossible to close, yet obscuring part of the text, specially on phones or tablets.

It is way worse on Chrome. In Firefox at least I can block some with ABP.

:) Chrome - don't touch it. chromium - only when I have to. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 18/12/2018 22.01, Carlos E. R. wrote:

On 18/12/2018 21.35, Dave Howorth wrote:

...

On Tue, 18 Dec 2018 20:34:10 +0100 "Carlos E. R." <> wrote:

...

On 18/12/2018 17.20, Dave Howorth wrote:

...

On Tue, 18 Dec 2018 15:26:31 +0100 Liam Proven wrote:

...

I opened a few tabs with a middle-click, almost at random.

#3 was this:

https://www.cnet.com/news/scientists-discover-rare-fossil-from-the-big-bang/

This has an autoplay video, unrelated to the story, at bottom right. Currently it is trying to tell me about "the top 5 best tech toys".

Hmm, there's no autoplay video on my machine, because cnet isn't allowed to run javascript. And if I do allow it for some reason, there's still nothing. If I further allow cbsistatic.com then there's a whole load of crap loads, but still no autoplay video. In fact no matter what I do there's no autoplay video, simply because I disabled it globally in Firefox's preferences. :)

I have disabled autoplay video, yet they play.

I've yet to encounter any example at all that does that. Could you provide a link to one? (Oh, you are using the opensuse packaged 60.3.0esr? Things are different in later versions as a quick search will explain)

It is not easy to find one now when I want to :-)

I did not mean that all videos play automatically. It is some on some sites with "clever" scripting, if I remember correctly they were commercials. I'll try [...] no luck.

https://www.dailymail.co.uk/news/article-6525515/The-worst-drivers-wheels-Ge... Going down the page there is a video, not started. As I go down past it, a float appears on the bottom right with the video, fortunately not started. And it can be closed. This time :-) But this one is not a commercial, but a part of the article. Commercials are more intrusive. -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)

On 18/12/2018 20:34, Carlos E. R. wrote:

I have disabled autoplay video, yet they play.

Exactly. I even tried an addon that claims to block all. It doesn't, it just gets a few more. Sometimes I just want to go back to nothing but plain-text email lists, Usenet and bulletin boards. But the email lists are full of top-posters. :-( -- Liam Proven - Technical Writer, SUSE Linux s.r.o. Corso II, Křižíkova 148/34, 186-00 Praha 8 - Karlín, Czechia Email: lproven@suse.com - Office telephone: +420 284 241 084 -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 2018-12-18 7:11 a.m., Liam Proven wrote:

Bookmark! Use bookmarks! Tabs are not bookmarks! Do not leave something open just because you intend to go back to it someday...

As in "yes it used to be but they changed all that". Yes, FF *USED* to open all and every tab on start-up. Now it doesn't. You have to visit the tab for it to open it. The same with Groups. Bookmarks are "file and forget". "Recent" gets pruned heavily (and its amazing how much junk FF seems to put in it) and cleared after a week. I can't recall the last time I retrieved something from "Bookmarks" other than cooking recipes. The same with incoming mail; I weekly move inbox deleted to 'old message's and them to 'archives' and then to storage archive. But then again, I'm very aggressive about deleting email. A week from not this thread will have gone POOF! And who cares? I can find it the on-line archives. I only save How-To and the like. -- A: Yes. > Q: Are you sure? >> A: Because it reverses the logical flow of conversation. >>> Q: Why is top posting frowned upon? -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 2018-12-18 9:41 a.m., Liam Proven wrote:

On 18/12/2018 13:44, Anton Aylward wrote:

...

Yes, FF *USED* to open all and every tab on start-up.

This is just a setting.

https://support.mozilla.org/en-US/questions/1016787

But you said you had hundreds of tabs open. I suggest that you don't want background-autoload if so -- otherwise re-opening Firefox will bog down your computer for several minutes while they all load!

Did I say "OPEN"? I've gone over my past posting and no, I didn't say that. In fact I described starting up with only the selected tab open. yes that is the setting I had, I wasn't even aware I could "start them all up" on bringing FF up. And yes it sounds a good way to bog the machine down. BTDT with chrome. -- A: Yes. > Q: Are you sure? >> A: Because it reverses the logical flow of conversation. >>> Q: Why is top posting frowned upon? -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 18/12/2018 18.55, Liam Proven wrote:

On 18/12/2018 17:23, Anton Aylward wrote:

...

Did I say "OPEN"? I've gone over my past posting and no, I didn't say that. In fact I described starting up with only the selected tab open. yes that is the setting I had, I wasn't even aware I could "start them all up" on bringing FF up.

Then I am sorry but I do not understand what you mean. :-(

Could you explain it to me?

Anton doesn't understand what we say of videos automatically playing on some unknown tab, because he has disabled the feature to load all tabs in background. Me too. Yet on some circumstances it happens, like what you say, opening several links at once from a mail with several links. But currently it doesn't happen when restoring a previous session, only one tab loads. -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)

On 19/12/2018 15:13, Anton Aylward wrote:

I also have a phone, running Android and there, Chrome. It is strictly a one-tab-at-a-time thing.

Try it in landscape mode. :-)

So I'm not ignorant of the phenomena in the abstract; it just doesn't happen for me on Firefox on the PC. Good for you!

Yes I have disabled the feature to load all tabs in background. Or perhaps it was delivered that way in some upgrade; I don't recall.

My mistake -- or one of them -- was that I thought you *wanted* this. -- Liam Proven - Technical Writer, SUSE Linux s.r.o. Corso II, Křižíkova 148/34, 186-00 Praha 8 - Karlín, Czechia Email: lproven@suse.com - Office telephone: +420 284 241 084 -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 19/12/2018 15:35, Anton Aylward wrote:

Why?

It's not an instruction, merely a suggestion!

I'm reading news. I want it in page mode so I can read more. I have no need for multiple tabs. The Flipboard finds the articles; I view the pages it finds in Chrome. Q.E.D. One at a time. Exit Chrome; scroll Flipboard for next interesting article. Only one Chrome tab ever needed. (Well, Chrome-beta usually)

I use tabs quite heavily in my mobile browsers, myself. -- Liam Proven - Technical Writer, SUSE Linux s.r.o. Corso II, Křižíkova 148/34, 186-00 Praha 8 - Karlín, Czechia Email: lproven@suse.com - Office telephone: +420 284 241 084 -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 2018-12-18 8:47 a.m., Carlos E. R. wrote:

On 18/12/2018 13.11, Liam Proven wrote:

...

On 17/12/2018 15:19, Anton Aylward wrote:

...

I had hoped that the Chromium model of one process per page would make this sort of thing more visible. But lets see, my FF has about 100 tabs (down from about 400 last month)

:-o

#Fear

I had many colleagues at RH who did this. It terrifies me. I currently have 8 and consider that a lot.

Then again, my personal inbox has 2 unread. I'd like to get back to zero soon. But my SO's has something like 14000 unread. I think I made a small terrified noise aloud when I installed an email client on her phone and saw the size of the number....

As I said to my colleagues:

Bookmark! Use bookmarks! Tabs are not bookmarks! Do not leave something open just because you intend to go back to it someday...

Bookmarks are insufficient.

Damn tooting right they are! it's not that they need a comment. Heck adding the one line comment in RCS was a pain, but for something like this there should be an automatic 'abstract' mechanism.

I also have hundreds of tabs. Now, what I would like is a page (could be another program) that displays (possibly frozen, or refreshed not fast, say once per hour) each tab in a small picture with readable text, so that I can remember what page has what I need so that I can open it. It would need to have some searchable text.

Agreed!

Now, "pocket" does something like that.

It does. I love pocket, but it is too easy to overload it. It is fine as an app on my tablet, but on my PC? It need a web page and that just buries the issue where we started. Can you say 'recursion'?

...

...

But Chromium handles some pages quite differently. If I visit Walmart.ca with FF I can't see prices, but I can if I use chromium.

I see prices in my FF:

https://www.walmart.ca/en/electronics/ipad-tablets/N-2428?icid=home%20page_C...

no, no prices displayed when I visit that with FF 52.8

https://www.walmart.ca/en/ip/rca-android-tablet-rct6973w43/6000197169006

Again, no, no price displayed. -- A: Yes. > Q: Are you sure? >> A: Because it reverses the logical flow of conversation. >>> Q: Why is top posting frowned upon? -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 2018-12-18 2:49 p.m., Carlos E. R. wrote:

...

...

Now, "pocket" does something like that.

...

It does. I love pocket, but it is too easy to overload it. It is fine as an app on my tablet, but on my PC? It need a web page and that just buries the issue where we started. Can you say 'recursion'?

There is a plugin or something, a tiny button "save to pocket".

Yes, I have that. It SAVES to pocket. If I want to SEE what is saved to pocket, the 'pin-board', that requires visiting a web page.

Another problem for me is that pocket loads the page from *their* location, not mine, so what I see in pocket is different. Sometimes a message that the content is not available for the USA, or totally different content. Sometimes it hits a "no robots allowed". And of course, some pages may need a login.

And yes, some sites are really USA not North America and you fan fall into the gap that Canada is North America (as is Mexico) but not the USA. And vice versa. The issue then isn't so much the can/can't as nation specific. And, OBTW: there are some items I see on eBay where a Chinese vendor will ship to the USA but not Canada. Go figure. What ever happened to 'internationalism'? -- A: Yes. > Q: Are you sure? >> A: Because it reverses the logical flow of conversation. >>> Q: Why is top posting frowned upon? -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 18/12/2018 15:09, Per Jessen wrote:

Wow. How do you guys manage those?

You may well ask. I saw many colleagues doing it at RH. I thought it was crazy. 2 tools help: * Tree Style Tabs https://addons.mozilla.org/en-US/firefox/addon/tree-style-tab/ * Tabhunter https://addons.mozilla.org/en-US/firefox/addon/tabhunter/ But I don't recommend the practice! -- Liam Proven - Technical Writer, SUSE Linux s.r.o. Corso II, Křižíkova 148/34, 186-00 Praha 8 - Karlín, Czechia Email: lproven@suse.com - Office telephone: +420 284 241 084 -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 18/12/2018 15.09, Per Jessen wrote:

Carlos E. R. wrote:

...

On 18/12/2018 13.11, Liam Proven wrote:

...

As I said to my colleagues:

Bookmark! Use bookmarks! Tabs are not bookmarks! Do not leave something open just because you intend to go back to it someday...

Bookmarks are insufficient.

I also have hundreds of tabs.

Wow. How do you guys manage those? I usually have 2-3 windows with maybe 10 tabs each. A window is usually "dedicated" for one type of thing/job/task.

Hundreds of tabs - with one process per tab, that must gobble a lot of memory?

4 processes max on FF, and only those I actually click do load. The rest display some information when hovering with the mouse. Many of the tabs are things that I have pending to do, or tabs that I open when I do certain things. For example, when I use Lazarus I open a bunch of tabs with different informations pertaining to that project. Another window may have tabs related to taxes. Whatever. Yes, I use those instead of bookmarks, but it is that bookmarks are very insufficient and take time to create properly, classify, and use. -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)

Anton Aylward composed on 2018-12-19 17:10 (UTC-0500):

On 2018-12-19 4:52 p.m., Carlos E. R. wrote:

...

Many of the tabs are things that I have pending to do, or tabs that I open when I do certain things.

I *always* have a tab for Google Search.

That would be redundant here: http://fm.no-ip.com/SS/Moz/ffSearchBox.jpg -- Evolution as taught in public schools is religion, not science. Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata *** http://fm.no-ip.com/ -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 2018-12-19 10:17 p.m., Carlos E. R. wrote:

On 19/12/2018 23.10, Anton Aylward wrote:

...

On 2018-12-19 4:52 p.m., Carlos E. R. wrote:

...

Many of the tabs are things that I have pending to do, or tabs that I open when I do certain things. I *always* have a tab for Google Search.

So? I'd have to remember the correct search terms, which are several, and then browse the findings to locate the correct results.

Far easier to have a window already with the 15 tabs ready.

Go back over what I said. I have all the tabs for my commonly visited sites already there; I just proliferate tabs without proliferating windows. I _used_ to proliferate groups, but there is something about groups and windows that seem to add to the load more than simply proliferating (unopened) tabs in one window, I've found). YMMV. -- A: Yes. > Q: Are you sure? >> A: Because it reverses the logical flow of conversation. >>> Q: Why is top posting frowned upon? -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 18/12/2018 15.42, Liam Proven wrote:

On 18/12/2018 14:47, Carlos E. R. wrote:

...

what did they say? I didn't notice.

Just a link suffices ;-)

https://www.theregister.co.uk/2018/12/06/microsoft_edge_chromium/

Thanks! [...] Wow. Very interesting. (further comments on opensuse-offtopic@opensuse.org if any) -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)

On 17/12/2018 09.46, David C. Rankin wrote:

On 12/16/2018 08:09 AM, Carlos E. R. wrote:

...

Well, no more crashes so far. Go figure. I will watch it, though.

Since yesterday night, the machine is running Thunderbird, Firefox, Chrome, and Kodi, no issues. Perfectly responsive with that large load.

Less than 3 GB of ram are used by processes, leaving half a gig free, another half for buffers/cache, and half available. This is the normal situation under "load". More than 1 gig of swap used.

Is the laptop old enough to begin worrying about puffy capacitors? Many time these can avoid the MCE checks and later cause havoc, the leave the laptop running fine for a few days until the humidity, pressure or temp varies, phase to the moon changes, etc.. and then the gremlin rears its ugly head again.

Not something easily checked on a laptop without major surgery. A desktop is a simple matter of popping the case and looking for puffy caps (the same applies to laptops, but the disassembly is much more involved...)

But this is a kernel issue, triggered by some unknown, not a hardware issue. -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)