[opensuse] Another example of deliberate systemd supporters deliberately sabotaging previously working software...
Just encountered this looking at /etc/sysconfig/sendmail: diff between last release and systemd -- all of the comments that listed the commands on how to do various sendmail maintenance functions were removed. They were comments!!! # diff /Nroot/etc/sysconfig/sendmail /etc/sysconfig/sendmail 7d6 < ## Command: /usr/lib/sendmail.d/update 17d15 < ## Command: /usr/lib/sendmail.d/update 41d38 .... 295d272 < ## Command: /usr/lib/sendmail.d/update 306d282 < ## Command: /usr/lib/sendmail.d/update --- There was no reason to change this file at all, but to make things more 'grand' -- they [werner Fink], didn't leave ".rpmold" or put his changes in .rpmnew -- they just edited (comment removal only that I saw in this case) them in place. --- the various sendmail rc scripts that did a good job of extracting options from the /etc/sendmail.cf config file -- replaced by "semi-hidden" (in /etc/mail/system -- not a standard location for storing service start/stop scripts -- except w/systemd. A wonderful benefit -- used to be the scripts were all in /etc/init.d/xxx but now, they can all be put in private, non-standard locations to obfuscate how systemd runs. I'm not one to rant on systemd other than for substantiated reasons like this. Again -- I will say this again -- the systemd people created a war with me by sabotage like this. If they had provided backward compat options... I would have loved them -- Many of their ideas were good ideas on Windows -- Win7, at least for 5-6 years and I've often tried to mention good features in windows that would make linux better (though I have more issues of problems in windows, and probably have more 'hate MS feelings', than their positive ideas have influenced me). Most of those hate feelings come from their manipulation of customers and the marketplace to force ideas down their throats. I don't like that behavior in MS or in linux, but now systemd supporters are adopting MS's behaviors. I know MS has done the embrace, extend, extinguish thing to lock in customers and force them onto "rental plans" -- end-users are the last to be converted to that. Adobe did it to their end users ~2 years ago and I disassociated myself with their community -- as they were all going on to monthly/yearly plans and would all have the newer SW that I couldn't buy. Big game companies are doing simpler with game clients that always have to be connected and have your system monitored. The two worst in that category - EA/Bioware/MassEffect3. I bought the game. I've never played the game I bought despite many calls to EA support -- but I have played versions with all the addons that had the anti-customer DRM removed. I never got to play online with other users, but I at least got all their extras for free -- and I still have the $60, legal/licensed copy that won't work. Another game recently -- also online -- after I downloaded 30G(!!!!) of textures -- I tried to login -- it wouldn't let me -- as it said that I had some "monitoring" software running that they considered cheating (they didn't say what the SW was, but in their help pages, I saw one of the largest offenders were real-time malware scanners (like Microsoft's Home Security Essentials). Real interesting. They were uninstalled in less than a day. So please realize that the harder you attempt to force your ideas of mandatory control on users -- the more users you will lose -- because of system problems -- who may still want to use your products -- but will only have the option of using the DRM-neutered ones that will actually work on their machines). *sad* Linda p.s. sorry for going 'off topic', but the stuff happening w/suse lately has really reminded me of the capitalist, bought and owned government that continues sell out to the highest bidder. It really tastes bitter. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On Sat, Aug 1, 2015 at 4:52 AM, Linda Walsh <suse@tlinx.org> wrote:
There was no reason to change this file at all, but to make things more 'grand' -- they [werner Fink], didn't leave ".rpmold" or put his changes in .rpmnew -- they just edited (comment removal only that I saw in this case) them in place.
What an intriguing horror story ... Werner Fink coming secretly in the night to your home and editing file without even leaving backup ... This is standard RPM behavior - if file did not change, there is nothing to preserve. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 On 2015-08-03 13:05, Andrei Borzenkov wrote:
On Sat, Aug 1, 2015 at 4:52 AM, Linda Walsh <> wrote:
This is standard RPM behavior - if file did not change, there is nothing to preserve.
I believe an rpm update does not backup scripts, only configuration files. Ie, local customizations. And not all customization: changes in scripts are lost. On the other hand, quite often in updates I have to compare both the old (saved by rpm command) and new files, and they were exactly the same. The rpm command should have fully replaced the old config file without the extra noise. But this is nothing done by systemd, unless Linda failed to explain the case better, instead of ranting ;-) - -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" (Minas Tirith)) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iF4EAREIAAYFAlW/X5UACgkQja8UbcUWM1wTbgEAoFiUFE8y0JpBe6QO2Lg6wjeW BQlQfjM8IGNIwhmAO/MA/1NqgDNykhoZ1bXNtmxIR37YhrS4A/sUuKc50/3eXI5K =aok7 -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Carlos E. R. wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256
On 2015-08-03 13:05, Andrei Borzenkov wrote:
On Sat, Aug 1, 2015 at 4:52 AM, Linda Walsh <> wrote:
This is standard RPM behavior - if file did not change, there is nothing to preserve.
I believe an rpm update does not backup scripts, only configuration files. Ie, local customizations. And not all customization: changes in scripts are lost.
It's anything that is marked as a config file in the rpm. I suppose it is possible that files in 'sysconfig' might not be considered config files...oh I forgot... I also keep many directories in '/etc', in RCS because I often make changes to those files -- want to preserve the suse originals as well as my changes. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On Mon, Aug 03, 2015 at 11:40:45AM -0700, Linda Walsh wrote:
Carlos E. R. wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256
On 2015-08-03 13:05, Andrei Borzenkov wrote:
On Sat, Aug 1, 2015 at 4:52 AM, Linda Walsh <> wrote:
This is standard RPM behavior - if file did not change, there is nothing to preserve.
I believe an rpm update does not backup scripts, only configuration files. Ie, local customizations. And not all customization: changes in scripts are lost.
It's anything that is marked as a config file in the rpm.
I suppose it is possible that files in 'sysconfig' might not be considered config files...oh I forgot... I also keep many directories in '/etc', in RCS because I often make changes to those files -- want to preserve the suse originals as well as my changes.
The /etc/sysconfig/files are changed by the fillup framework, using templates in /var/adm/fillup-templates/sysconfig.* The fillup framework is however careful not to break anything. Ciao, Marcus -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 08/04/2015 05:06 AM, Marcus Meissner wrote:
The /etc/sysconfig/files are changed by the fillup framework, using templates in /var/adm/fillup-templates/sysconfig.*
Wow, thank you! More RTFM. Still learning after all these years. -- 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
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 Resending because of "moderation" with altered subject line. Apparently, Andrei messages are not delayed for moderation, but mine are :-/ On 2015-08-03 13:05, Andrei Borzenkov wrote:
On Sat, Aug 1, 2015 at 4:52 AM, Linda Walsh <> wrote:
This is standard RPM behavior - if file did not change, there is nothing to preserve.
I believe an rpm update does not backup scripts, only configuration files. Ie, local customizations. And not all customization: changes in scripts are lost. On the other hand, quite often in updates I have to compare both the old (saved by rpm command) and new files, and they were exactly the same. The rpm command should have fully replaced the old config file without the extra noise. But this is nothing done by systemd, unless Linda failed to explain the case better, instead of ranting ;-) - -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" (Minas Tirith)) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iF4EAREIAAYFAlW/YOcACgkQja8UbcUWM1w6jwD/c3gcKBA42Dbauw/ysnI4W0Ub i9SAAd5uzjw0rrMUgyIA/28wHQmfSkHJva1hnOniWQkveCnC47mfxHHzzeq/wLv2 =WtXL -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 08/03/2015 08:39 AM, Carlos E. R. wrote:
But this is nothing done by systemd, unless Linda failed to explain the case better, instead of ranting ;-)
+1 -- 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 Mon, Aug 3, 2015 at 3:39 PM, Carlos E. R. <robin.listas@telefonica.net> wrote:
On 2015-08-03 13:05, Andrei Borzenkov wrote:
On Sat, Aug 1, 2015 at 4:52 AM, Linda Walsh <> wrote:
This is standard RPM behavior - if file did not change, there is nothing to preserve.
I believe an rpm update does not backup scripts, only configuration files. Ie, local customizations. And not all customization: changes in scripts are lost.
I understood it so that reference to update directory was commented out in /etc/sysconfig/sendmail and Linda complained that previous version of /etc/sysconfig file was not preserved. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 On 2015-08-03 15:15, Andrei Borzenkov wrote:
On Mon, Aug 3, 2015 at 3:39 PM, Carlos E. R. <> wrote:
I understood it so that reference to update directory was commented out in /etc/sysconfig/sendmail and Linda complained that previous version of /etc/sysconfig file was not preserved.
I believe they never are. Perhaps they should. - -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" (Minas Tirith)) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iF4EAREIAAYFAlW/bk4ACgkQja8UbcUWM1xJXAD9H+ubabb9h94T6nSNtRzYgUq9 iw9IL0/AcsRqjnuhbaEBAIZKdLnWwCoAC90gGWzsAuZf66h0g6cpiHSIGpeDbUi6 =Hdwa -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 08/03/2015 09:36 AM, Carlos E. R. wrote:
On 2015-08-03 15:15, Andrei Borzenkov wrote:
On Mon, Aug 3, 2015 at 3:39 PM, Carlos E. R. <> wrote:
I understood it so that reference to update directory was commented out in /etc/sysconfig/sendmail and Linda complained that previous version of /etc/sysconfig file was not preserved.
I believe they never are. Perhaps they should.
Perhaps if she was using that other radically new, innovative things that has taken almost as much abuse and derision as systemd: BtrFS and snapshots, it would have been preserved. Oh, wait! Doesn't she advocate XFS? Can't that do snapshots like BtrFS, like ext4? Well, OK, perhaps not as smooth and well integrated as BtrFS .... Oh wait! When I decide that I would prefer the old version I just use zypper (or yast) to go back to a previous version. its out there somewhere ... -- 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
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 On 2015-08-03 16:24, Anton Aylward wrote:
On 08/03/2015 09:36 AM, Carlos E. R. wrote:
I believe they never are. Perhaps they should.
Perhaps if she was using that other radically new, innovative things that has taken almost as much abuse and derision as systemd: BtrFS and snapshots, it would have been preserved.
Certainly. Yast/zypper does a snapshot before and after an update, I understand.
Oh, wait! Doesn't she advocate XFS?
So do I :-)
Can't that do snapshots like BtrFS, like ext4?
No, or not currently. Actually, XFS shares some devs and features with btrfs.
Oh wait! When I decide that I would prefer the old version I just use zypper (or yast) to go back to a previous version. its out there somewhere ...
Many do a local backup of the entire etc directory ;-) - -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" (Minas Tirith)) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iF4EAREIAAYFAlW/fH4ACgkQja8UbcUWM1wdBgEAmWTmRvuGGsgq4S2dGKhKe37N ctXqyOFJ6gGdnds1B3EA/1Q3H3PFC4GaGEkVNoqCSGsga1oH7qJ7XB0iKL0tCSAx =CaXq -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 08/03/2015 10:36 AM, Carlos E. R. wrote:
Can't that do snapshots like BtrFS, like ext4? No, or not currently. Actually, XFS shares some devs and features with btrfs.
I recall reading that XFS could do snapshots but required (a) not to mount via uuid and (b) a "freeze" option. I also recall that the freeze option became kernel sported and could even be used with ext3, but I don't recall where I read that. -- 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
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 On 2015-08-03 16:44, Anton Aylward wrote:
On 08/03/2015 10:36 AM, Carlos E. R. wrote:
Can't that do snapshots like BtrFS, like ext4? No, or not currently. Actually, XFS shares some devs and features with btrfs.
I recall reading that XFS could do snapshots but required (a) not to mount via uuid and (b) a "freeze" option. I also recall that the freeze option became kernel sported and could even be used with ext3, but I don't recall where I read that.
The name is the same, but the feature is distinctly different. The purpose is to create a backup of the filesystem into a different device. It is not a COW (copy on write). - -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" (Minas Tirith)) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iF4EAREIAAYFAlW/fzgACgkQja8UbcUWM1z0PAEAkV7Xjc8DOxYj4WLgoGEgFnCJ P7wuJ7dsk0NUgHxcT4QA/1t3LZGhR+P5/bdzJeGJHFgXYDotSWV8b3uKdjxX9URS =gyh+ -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 08/03/2015 10:48 AM, Carlos E. R. wrote:
The name is the same, but the feature is distinctly different.
The purpose is to create a backup of the filesystem into a different device. It is not a COW (copy on write).
So it really generational backup for potential offlining, perhaps like disk-to-disk-to-tape. -- 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 2015-08-03 16:54, Anton Aylward wrote:
On 08/03/2015 10:48 AM, Carlos E. R. wrote:
The name is the same, but the feature is distinctly different.
The purpose is to create a backup of the filesystem into a different device. It is not a COW (copy on write).
So it really generational backup for potential offlining, perhaps like disk-to-disk-to-tape.
From memory, and maybe faulty because I have never done it. I think the procedure is to block changes to the filesystem (this is the part I'm unsure about), then clone or image the filesystem somewhere else, as fast as possible, then free up changes again. If the filesystem would be able to freeze changes, but still accept them, written somewhere else, delayed, then applied later when the freeze is removed, that would be wonderful. I don't know if XFS is capable of this. Take a look at xfsdump and xfs_copy. xfsdump creates an image of an XFS filesystem, and does it very fast. It creates a single target file (an archive). It can be incremental. You do not need to umount it, I believe (I don't remember for sure; need more coffe). xfs_copy instead copies trees of files into another tree. Source and target must be XFS filesystems. Source should be be umounted, but could be "frozen" instead. It is similar to using rsync, but runs much faster, because it takes advantage of knowledge of the metadata structure. xfs_freeze - suspend access to an XFS filesystem The manual mentions that is intended for use with lvm or raid with snapshot support, but I don't see how. -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 В Mon, 3 Aug 2015 16:48:24 +0200 "Carlos E. R." <robin.listas@telefonica.net> пишет:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256
On 2015-08-03 16:44, Anton Aylward wrote:
On 08/03/2015 10:36 AM, Carlos E. R. wrote:
Can't that do snapshots like BtrFS, like ext4? No, or not currently. Actually, XFS shares some devs and features with btrfs.
I recall reading that XFS could do snapshots but required (a) not to mount via uuid and (b) a "freeze" option. I also recall that the freeze option became kernel sported and could even be used with ext3, but I don't recall where I read that.
The name is the same, but the feature is distinctly different.
The purpose is to create a backup of the filesystem into a different device. It is not a COW (copy on write).
Actually it is exactly what CoW is. It is btrfs which is not CoW - more appropriate name is "redirect on write" or "write anywhere" (which is how oldest implementation is called). There is no "copy" involved in btrfs write process. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iEYEARECAAYFAlW/lVEACgkQR6LMutpd94xybQCgr5FzuGHPh90lZ+bUjesNOa+6 QiIAoLP7RcERcHe3wG6ExSAYLxmtSlIC =T620 -----END PGP SIGNATURE-----
On 2015-08-03 18:22, Andrei Borzenkov wrote:
В Mon, 3 Aug 2015 16:48:24 +0200 "Carlos E. R." <> пишет:
Actually it is exactly what CoW is. It is btrfs which is not CoW - more appropriate name is "redirect on write" or "write anywhere" (which is how oldest implementation is called).
There is no "copy" involved in btrfs write process.
Ah. I still don't have those names correct. :-) -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)
On Mon, Aug 3, 2015 at 10:22 AM, Andrei Borzenkov <arvidjaar@gmail.com> wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
В Mon, 3 Aug 2015 16:48:24 +0200 "Carlos E. R." <robin.listas@telefonica.net> пишет:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256
On 2015-08-03 16:44, Anton Aylward wrote:
On 08/03/2015 10:36 AM, Carlos E. R. wrote:
Can't that do snapshots like BtrFS, like ext4? No, or not currently. Actually, XFS shares some devs and features with btrfs.
I recall reading that XFS could do snapshots but required (a) not to mount via uuid and (b) a "freeze" option. I also recall that the freeze option became kernel sported and could even be used with ext3, but I don't recall where I read that.
The name is the same, but the feature is distinctly different.
The purpose is to create a backup of the filesystem into a different device. It is not a COW (copy on write).
Actually it is exactly what CoW is. It is btrfs which is not CoW - more appropriate name is "redirect on write" or "write anywhere" (which is how oldest implementation is called).
There is no "copy" involved in btrfs write process.
COW semantically means deduplication, i.e. when multiple functions want resources that are indistinguishable, COW is the optimization that permits pointing all functions to a single copy of that resource. So it doesn't mean "when writing, duplicate, then write". It's originally a memory optimization. For file systems it means effectively "do not overwrite on write". And therefore Btrfs (and ZFS) is legitimately called a COW file system. Where XFS and ext234 do overwrites of existing files. On Btrfs, this extends to file system metadata too, with exception of superblocks, it's not overwritten when there are changes, even deletions. So a file delete involves a new write to free space, then walking back through the tree to remove blocks making them free space again as needed. As for freeze. I think this was originally an XFS invention, the gist is that it commits all pending writes to stable media and then makes the file system look cleanly unmounted while not unmounting it. That way an *LVM* snapshot can be taken, then the original is unfrozen and write can continue to happen, while the snapshot is used for doing backups that are consistent and unchanging. But now these days fsfreeze uses the FIFREEZE ioctl which also lets it work for ext4, and then some time ago there were patches to make it work for Btrfs also. https://lwn.net/Articles/379862/ on the original btrfs patches with explanation; patches have evolved in the actual kernel code https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/fs/supe... starting at line 1258 has comments on the freezing stages. and also https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/fs/bloc... starting at line 200 has comments on how the block device status is handled As for systemd problems, I'm definitely missing the emails that explain what the manifestation is. But I think it's understandable to be frustrated when confused, and it's understandable to be confused with significant changes like what systemd implements. Linux stuff is like any big city in the world that has good parts and stinky parts and always parts under some state of deconstruction and construction, I see no problem with expressing annoyance with such things, even in the face of futility. But it's a.) not in evidence and b.) not on topic for this list, to assign willful sabotage of peoples' workflows to particular development projects. *shrug* -- Chris Murphy -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Oh and XFS does not support snapshotting. This is done via LVM. The practice is to freeze the file system, then snapshot, then unfreeze. I forget if the freeze is done by LVM these days automatically or not; it seems like a good optimization, I'm not sure in what instance you would not want the fs frozen first? --- Chris Murphy -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
В Mon, 3 Aug 2015 12:33:56 -0600 Chris Murphy <lists@colorremedies.com> пишет:
Actually it is exactly what CoW is. It is btrfs which is not CoW - more appropriate name is "redirect on write" or "write anywhere" (which is how oldest implementation is called).
There is no "copy" involved in btrfs write process.
COW semantically means deduplication,
"When I use a word," Humpty Dumpty said, in rather a scornful tone, "it means just what I choose it to mean—neither more nor less." Please, please - CoW snapshot means - before new data is WRITTEN, original data is COPIED somewhere else. CoW snapshots have absolutely nothing to do with deduplication and existed long before deduplication started to be used in storage world. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On Mon, Aug 3, 2015 at 12:51 PM, Andrei Borzenkov <arvidjaar@gmail.com> wrote:
В Mon, 3 Aug 2015 12:33:56 -0600 Chris Murphy <lists@colorremedies.com> пишет:
Actually it is exactly what CoW is. It is btrfs which is not CoW - more appropriate name is "redirect on write" or "write anywhere" (which is how oldest implementation is called).
There is no "copy" involved in btrfs write process.
COW semantically means deduplication,
"When I use a word," Humpty Dumpty said, in rather a scornful tone, "it means just what I choose it to mean—neither more nor less."
Please, please - CoW snapshot means - before new data is WRITTEN, original data is COPIED somewhere else.
No, that's just how LVM thick did it. LVM thin does it differently, and it's COW.
CoW snapshots have absolutely nothing to do with deduplication and existed long before deduplication started to be used in storage world.
You're just thinking of this in terms of the LVM thick implementation, not the long standing denotation of the term. It was something of an evil necessity that LVM did it the way it did it, and that's why it doesn't perform well and will eventually be supplanted by LVM thin snapshots. -- Chris Murphy -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On August 3, 2015 3:11:53 PM EDT, Chris Murphy <lists@colorremedies.com> wrote:
On Mon, Aug 3, 2015 at 12:51 PM, Andrei Borzenkov <arvidjaar@gmail.com> wrote:
В Mon, 3 Aug 2015 12:33:56 -0600 Chris Murphy <lists@colorremedies.com> пишет:
Actually it is exactly what CoW is. It is btrfs which is not CoW -
more
appropriate name is "redirect on write" or "write anywhere" (which is how oldest implementation is called).
There is no "copy" involved in btrfs write process.
COW semantically means deduplication,
"When I use a word," Humpty Dumpty said, in rather a scornful tone, "it means just what I choose it to mean—neither more nor less."
Please, please - CoW snapshot means - before new data is WRITTEN, original data is COPIED somewhere else.
No, that's just how LVM thick did it. LVM thin does it differently, and it's COW.
CoW snapshots have absolutely nothing to do with deduplication and existed long before deduplication started to be used in storage world.
You're just thinking of this in terms of the LVM thick implementation, not the long standing denotation of the term. It was something of an evil necessity that LVM did it the way it did it, and that's why it doesn't perform well and will eventually be supplanted by LVM thin snapshots.
Got a link describing LVM thin snapshots? I thought they were just traditional snapshots applied to thin provisioned volumes. Greg -- Sent from my Android device with K-9 Mail. Please excuse my brevity. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Anton Aylward wrote:
On 08/03/2015 10:36 AM, Carlos E. R. wrote:
Can't that do snapshots like BtrFS, like ext4? No, or not currently. Actually, XFS shares some devs and features with btrfs.
I recall reading that XFS could do snapshots but required (a) not to mount via uuid and (b) a "freeze" option. I also recall that the freeze option became kernel sported and could even be used with ext3, but I don't recall where I read that.
3rd hand gossip -- they just implemented a non-default option to do store a checksum of metadata if you also used the new features to keep around a special extent for free space. the GUID, apparently, was also part of the "metadata" that was crc'd. (BTW, I had a cow when I heard this)...but a _patch_ was submitted within the last week to move the user-settable GUID out of the crc'd data...). But I still don't like being forced to have crc'd metadata just to get the free-space extents -- they said it was needed for reliability -- which may be true, dunno, it sounded like an "agenda-ist" excuse... i.e. something that doesn't make technical sense to *force* on users. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Anton Aylward wrote:
On 08/03/2015 09:36 AM, Carlos E. R. wrote:
On 2015-08-03 15:15, Andrei Borzenkov wrote:
On Mon, Aug 3, 2015 at 3:39 PM, Carlos E. R. <> wrote: I understood it so that reference to update directory was commented out in /etc/sysconfig/sendmail and Linda complained that previous version of /etc/sysconfig file was not preserved. I believe they never are. Perhaps they should.
Perhaps if she was using that other radically new, innovative things that has taken almost as much abuse and derision as systemd: BtrFS and snapshots, it would have been preserved.
Oh, wait! Doesn't she advocate XFS? Can't that do snapshots like BtrFS, like ext4?
my HOME partition is xfs and ... and, well, until some breakage happened that I haven't had time to track down occurred.... had daily snapshots. As far as going back to previously working copies... I had a file that disappeared 3 months ago... So I restored it from backups. BtrFS seems to be causing disk full crashes within weeks to days with some people -- and now you want me to believe I could have restored it from a 3mo-old backup?
Well, OK, perhaps not as smooth and well integrated as BtrFS ....
As for the file not being preserved -- that was on my active file system -- none of the several root+usr+var backups were affected. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Carlos E. R. wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256
Resending because of "moderation" with altered subject line. Apparently, Andrei messages are not delayed for moderation, but mine are :-/
On 2015-08-03 13:05, Andrei Borzenkov wrote:
On Sat, Aug 1, 2015 at 4:52 AM, Linda Walsh <> wrote:
This is standard RPM behavior - if file did not change, there is nothing to preserve.
"my comments" about what command I might use to do these functions manually were deleted. How is that not a change?
I believe an rpm update does not backup scripts, only configuration files. Ie, local customizations. And not all customization: changes in scripts are lost.
---- /etc/sysconfig/sendmail is a config file -- it is also sourced by scripts.
On the other hand, quite often in updates I have to compare both the old (saved by rpm command) and new files, and they were exactly the same. The rpm command should have fully replaced the old config file without the extra noise.
--- I've seen that happen if the time/date stamp was not the same.. You don't have to change the comments.
But this is nothing done by systemd, unless Linda failed to explain the case better, instead of ranting ;-)
Never said it was -- said sd supporters... taking out the knowledge and/or means for people to do things the pre-sd way, isn't necessary to convert to system -- only to remove the ability to do it any other way. RANT RANT RANT!! :plplplbbbbbbbb: -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 2015-08-03 20:17, Linda Walsh wrote:
Carlos E. R. wrote:
I believe an rpm update does not backup scripts, only configuration files. Ie, local customizations. And not all customization: changes in scripts are lost.
/etc/sysconfig/sendmail is a config file -- it is also sourced by scripts.
Well, I'm afraid that files in that directory are not copied to a backup by rpm when changed. I don't know why. But this is not related at all to systemd doings. It is rather the sendmail maintainers people doings :-p -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)
В Mon, 03 Aug 2015 11:17:54 -0700 Linda Walsh <suse@tlinx.org> пишет:
Carlos E. R. wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256
Resending because of "moderation" with altered subject line. Apparently, Andrei messages are not delayed for moderation, but mine are :-/
On 2015-08-03 13:05, Andrei Borzenkov wrote:
On Sat, Aug 1, 2015 at 4:52 AM, Linda Walsh <> wrote:
This is standard RPM behavior - if file did not change, there is nothing to preserve.
"my comments" about what command I might use to do these functions manually were deleted.
How is that not a change?
My mistake - as I see it now, /etc/sysconfig is managed differently in *SUSE. Files are not part of RPM, rather they are generated during installation from templates with fillup command. So usual RPM configuration file logic does not apply here.
I believe an rpm update does not backup scripts, only configuration files. Ie, local customizations. And not all customization: changes in scripts are lost.
---- /etc/sysconfig/sendmail is a config file -- it is also sourced by scripts.
Yes, but apparently files in /etc/sysconfig are not meant to be manually edited and have rather strict structure. You should verify whether your changes comply with what /bin/fillup expects. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On Mon, Aug 3, 2015 at 12:17 PM, Linda Walsh <suse@tlinx.org> wrote:
Never said it was -- said sd supporters... taking out the knowledge and/or means for people to do things the pre-sd way, isn't necessary to convert to system -- only to remove the ability to do it any other way.
If you mean compatibility with sysvinit scripts, then even before this page was written up it was expected the effort to maintain it was going to be dropped on the systemd side; which would presumably leave it up to distros if they wanted to maintain such compatibility (which is the case to a great degree on CentOS 7). http://www.freedesktop.org/wiki/Software/systemd/Incompatibilities/ I think in 25 years when there is a new wizbang replacement for systemd, there will be ranting for the good old days of systemd and how it's appalling and wrong that backward compatibility not be maintained. I might be a complainer for all I know. -- Chris Murphy -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 08/03/2015 02:47 PM, Chris Murphy wrote:
I think in 25 years when there is a new wizbang replacement for systemd, there will be ranting for the good old days of systemd and how it's appalling and wrong that backward compatibility not be maintained. I might be a complainer for all I know.
In 25 years? Perhaps the rate of change is accelerating, perhaps it will be less than that. As for ranting, yes you will. I might too. Tempora mutantur, nos et mutamur in illis, -- 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
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 On 2015-08-01 03:52, Linda Walsh wrote:
Just encountered this looking at /etc/sysconfig/sendmail:
diff between last release and systemd -- all of the comments that listed the commands on how to do various sendmail maintenance functions were removed. They were comments!!!
# diff /Nroot/etc/sysconfig/sendmail /etc/sysconfig/sendmail 7d6 < ## Command: /usr/lib/sendmail.d/update 17d15 < ## Command: /usr/lib/sendmail.d/update 41d38 .... 295d272 < ## Command: /usr/lib/sendmail.d/update 306d282 < ## Command: /usr/lib/sendmail.d/update
---
You fail to prove any change, both sections are exactly the same. :-? - -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" (Minas Tirith)) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iF4EAREIAAYFAlW/XjAACgkQja8UbcUWM1xjcgD/S34LYQyK4B87SC/kEVMix9mi 3WwRWRK68f75vx6BzzsA/0QRYV9Es7XJH0sPRd3rvegTL8FkLSu6lfKPSAWV7nx/ =0cgP -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Carlos, et al I finally get to participate in a systemd flame tirade! :-) ...and then Carlos E. R. said... % ... % > # diff /Nroot/etc/sysconfig/sendmail /etc/sysconfig/sendmail 7d6 < % > ## Command: /usr/lib/sendmail.d/update 17d15 < ## Command: % > /usr/lib/sendmail.d/update 41d38 .... 295d272 < ## Command: % > /usr/lib/sendmail.d/update 306d282 < ## Command: % > /usr/lib/sendmail.d/update % % You fail to prove any change, both sections are exactly the same. :-? Not quite... If you unwrap this, you get < ## Command: ... < ## Command: ... ... but no corresponding
new content here
section. It would probably be more evident with a unified diff diff -bu /Nroot/etc/sysconfig/sendmail /etc/sysconfig/sendmail that the /Nroot/etc copy has the lines while the /etc copy doesn't. % % - -- % Cheers / Saludos, HTH & HAND :-D -- David T-G See http://justpickone.org/davidtg/email/ See http://justpickone.org/davidtg/tofu.txt -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 On 2015-08-03 14:32, David T-G wrote:
section. It would probably be more evident with a unified diff
Well, that's what I say, that I can not see what was changed. - -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" (Minas Tirith)) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iF4EAREIAAYFAlW/X+sACgkQja8UbcUWM1yGIQD8DtDq0utSlvMcvEbxuO77QS0h A0H9KCSGDpEvVAex+okA/jcnyJR6Dg4PzoVYPiAVsnmBt3YbuEh5NpX6bX0qeA72 =4c1n -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Carlos, et al -- ...and then Carlos E. R. said... % % -----BEGIN PGP SIGNED MESSAGE----- % Hash: SHA256 % % On 2015-08-03 14:32, David T-G wrote: % > section. It would probably be more evident with a unified diff % % Well, that's what I say, that I can not see what was changed. Lines were there in the /Nroot/etc version. Lines are missing in the /etc version. The change was to delete (not modify) those four lines. % % - -- % Cheers / Saludos, % % Carlos E. R. HTH & HAND :-D -- David T-G See http://justpickone.org/davidtg/email/ See http://justpickone.org/davidtg/tofu.txt -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 08/03/2015 08:40 AM, David T-G wrote:
The change was to delete (not modify) those four lines.
Aka stripping out comment 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
Anton, et al -- ...and then Anton Aylward said... % % On 08/03/2015 08:40 AM, David T-G wrote: % > % > The change was to delete (not modify) those four lines. % % Aka stripping out comment line? Yep. HAND :-D -- David T-G See http://justpickone.org/davidtg/email/ See http://justpickone.org/davidtg/tofu.txt -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
David T-G wrote:
Anton, et al --
...and then Anton Aylward said... % % On 08/03/2015 08:40 AM, David T-G wrote: % > % > The change was to delete (not modify) those four lines. % % Aka stripping out comment line?
Wrong. It is a comment line only to those who look at it as a shell-script input file, but if you ever used yast2 to change system settings, those were config lines that told yast2 how to effect the changes you wanted without having to reboot your system. Geez, people seem to forget how their systems used to function very quickly. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 On 2015-08-03 14:40, David T-G wrote:
Carlos, et al --
...and then Carlos E. R. said... % % -----BEGIN PGP SIGNED MESSAGE----- % Hash: SHA256
Er... please change your quote character to ">". Your "%" is not standard and thus not detected by MUAs.
Lines were there in the /Nroot/etc version. Lines are missing in the /etc version.
The change was to delete (not modify) those four lines.
Yes, I know that. But what exactly was in those lines? - -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" (Minas Tirith)) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iF4EAREIAAYFAlW/Y2UACgkQja8UbcUWM1zrqwEAgR3hdtbuubC4EjhkobBiUjw1 gycXn39Q/V70A1t3A4gA/3fhbcc71RraDpaGNX4xVZo5p4LEpd1KF8Gnl+zWaMVT =d7Fo -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Carlos, et al -- This seems to have become a fork of its own. ...and then Carlos E. R. said... % % On 2015-08-03 14:40, David T-G wrote: % > Carlos, et al -- % > % > ...and then Carlos E. R. said... % % -----BEGIN PGP SIGNED % > MESSAGE----- % Hash: SHA256 % % Er... please change your quote character to ">". Your "%" is not % standard and thus not detected by MUAs. To be brief, no :-) I'd be happy to go into detail if anyone is interested, but probably not. % % > Lines were there in the /Nroot/etc version. Lines are missing in % > the /etc version. % > % > The change was to delete (not modify) those four lines. % % Yes, I know that. But what exactly was in those lines? The ## Comment: ... ## Comment: ... content that you see in the "before" section. % % - -- % Cheers / Saludos, % % Carlos E. R. HTH & HAND :-D -- David T-G See http://justpickone.org/davidtg/email/ See http://justpickone.org/davidtg/tofu.txt -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 On 2015-08-03 14:55, David T-G wrote:
Carlos, et al --
% Er... please change your quote character to ">". Your "%" is not % standard and thus not detected by MUAs.
To be brief, no :-) I'd be happy to go into detail if anyone is interested, but probably not.
Please reconsider. You are breaking mail for the rest of the world.
The
## Comment: ... ## Comment: ...
content that you see in the "before" section.
Again, I want to see the exact and complete text that was altered. Not a saying that text was altered. - -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" (Minas Tirith)) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iF4EAREIAAYFAlW/ZZAACgkQja8UbcUWM1z8ggD/TsQHmOUw598NEYcow5xvWwdx OMswyCclvO9tZEcnIcIA/jKy/kMDinlzBZ3Y5znlNuYRmyZ9LJcx5fOxW0on6HMl =PdIc -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Carlos, et al -- ...and then Carlos E. R. said... % % On 2015-08-03 14:55, David T-G wrote: % > Carlos, et al -- % % > % Er... please change your quote character to ">". Your "%" is not % > % standard and thus not detected by MUAs. % > % > To be brief, no :-) I'd be happy to go into detail if anyone is % > interested, but probably not. % % Please reconsider. You are breaking mail for the rest of the world. That's a bit much. You and I have both been here for a long time, and I've been using %_ since the 90s. Enjoy http://grokbase.com/p/php/php-general/039jzxfvps/php-webhost-0t for a grin :-) % % > The % > % > ## Comment: ... ## Comment: ... % > % > content that you see in the "before" section. % % Again, I want to see the exact and complete text that was altered. Not % a saying that text was altered. The original text was in Linda's email; I'm just too lazy to retype it all, especially when your MUA reformatted the text as though it were a simple paragraph. I guess you're just going to have to get the two files in question and diff them for yourself. Perhaps Linda will send them to you. I don't know any other way to explain it. % % - -- % Cheers / Saludos, % % Carlos E. R. HAND :-D -- David T-G See http://justpickone.org/davidtg/email/ See http://justpickone.org/davidtg/tofu.txt -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 08/03/2015 09:05 AM, David T-G wrote:
The original text was in Linda's email;
Linda's 'original' was the output of a diff in edit mode showing 4 lines had been deleted. It was not the complete text that Carlos asked for. -- 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
Anton, et al -- ...and then Anton Aylward said... % % On 08/03/2015 09:05 AM, David T-G wrote: % > The original text was in Linda's email; % % Linda's 'original' was the output of a diff in edit mode showing 4 lines % had been deleted. It was not the complete text that Carlos asked for. Oh! I see. Thanks for the clarification! HAND :-D -- David T-G See http://justpickone.org/davidtg/email/ See http://justpickone.org/davidtg/tofu.txt -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 On 2015-08-03 15:21, Anton Aylward wrote:
On 08/03/2015 09:05 AM, David T-G wrote:
The original text was in Linda's email;
Linda's 'original' was the output of a diff in edit mode showing 4 lines had been deleted. It was not the complete text that Carlos asked for.
Exactly. - -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" (Minas Tirith)) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iF0EAREIAAYFAlW/brgACgkQja8UbcUWM1x7kQD+PgGYeDkjS4t7qmkt9DXC8xM6 l+MKrmqaiUjf2d3gFoIA91MJc2XdgRfhdWJvPFfRvwYfJEaCyNBVG9YZ+XRMUNg= =5Z+8 -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Anton Aylward wrote:
On 08/03/2015 09:05 AM, David T-G wrote:
The original text was in Linda's email;
Linda's 'original' was the output of a diff in edit mode showing 4 lines had been deleted. It was not the complete text that Carlos asked for.
I elided 20 other instances..., thus the .... in the middle, but yeah, I'd hoped that the "diff xxxx yyyy" at the beginning would clue some people in... but these days people 'HAND' and +1 each other over their own ignorance. You've heard the politico-social version of this probably: “if you repeat falsehoods long enough and over long enough time, ‘the people’ start thinking of the falsehood as true.” -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 08/03/2015 02:57 PM, Linda Walsh wrote:
Anton Aylward wrote:
On 08/03/2015 09:05 AM, David T-G wrote:
The original text was in Linda's email;
Linda's 'original' was the output of a diff in edit mode showing 4 lines had been deleted. It was not the complete text that Carlos asked for.
I elided 20 other instances..., thus the .... in the middle,
Yes right, you elided ... so it wasn't the complete text that Carlos asked for. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 08/03/2015 09:05 AM, David T-G wrote:
I'm just too lazy to retype it
No need to. Just scroll back to Linda's message and cut-and-paste. its only a few lines. Its only the output of a diff, and quite possibly not even all of that. Its not as if she posted the complete body of both files. -- 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
Anton, et al -- ...and then Anton Aylward said... % % On 08/03/2015 09:05 AM, David T-G wrote: % > I'm just too lazy to retype it % % No need to. % Just scroll back to Linda's message and cut-and-paste. Scroll? Cut? What is this magick of which you speek? I'm on a no-mouse xterm emulator at the moment :-) % its only a few lines. Its only the output of a diff, and quite % possibly not even all of that. Its not as if she posted the complete % body of both files. Agreed. I could even have used Carlos's quoted text to un-reformat it, but I just didn't want to invest that much time in it. When I thought he simply saw < < and wanted to see
because it's missing, I thought I'd clarify. Now that you have rightly realized that he in fact wants to see the whole source, we realize that I wouldn't be able to help him anyway. Ah, well. And so goes my first foray into the systemd pool :-) HAND :-D -- David T-G See http://justpickone.org/davidtg/email/ See http://justpickone.org/davidtg/tofu.txt -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 08/03/2015 09:59 AM, David T-G wrote:
Ah, well. And so goes my first foray into the systemd pool
Shame it had to come about as a result of one of Linda's rants :-( -- 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
Anton, et al -- ...and then Anton Aylward said... % % On 08/03/2015 09:59 AM, David T-G wrote: % > Ah, well. And so goes my first foray into the systemd pool % % Shame it had to come about as a result of one of Linda's rants :-( Couldn't be better, actually... I have no experience and thus nothing of significance to contribute to the systemd discussion, so it might as well be something as fluffy and pointless as replying to a rant and muddying what I was trying to clarify :-) HAND :-D -- David T-G See http://justpickone.org/davidtg/email/ See http://justpickone.org/davidtg/tofu.txt -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
David T-G wrote:
Carlos, et al
I finally get to participate in a systemd flame tirade! :-)
---- If you read all of my original -- you'd see it WASN'T a sd tirade, but an example about the move to sd has been *forced*, without leaving any choice to user's about how they want to manage their systems. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 08/03/2015 02:43 PM, Linda Walsh wrote:
David T-G wrote:
Carlos, et al
I finally get to participate in a systemd flame tirade! :-)
----
If you read all of my original -- you'd see it WASN'T a sd tirade, but an example about the move to sd has been *forced*, without leaving any choice to user's about how they want to manage their systems.
DUH? And that is different from a systemd tirade how, exactly? Oh, right, it was really an anti-Microsoft (or possibly an anti-establishment) tirade :-) -- 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 Mon, Aug 3, 2015 at 2:37 PM, Anton Aylward <opensuse@antonaylward.com> wrote:
On 08/03/2015 02:43 PM, Linda Walsh wrote:
David T-G wrote:
Carlos, et al
I finally get to participate in a systemd flame tirade! :-)
----
If you read all of my original -- you'd see it WASN'T a sd tirade, but an example about the move to sd has been *forced*, without leaving any choice to user's about how they want to manage their systems.
DUH? And that is different from a systemd tirade how, exactly?
I'm pretty sure a tirade must be long. While this thread is long, and has fragmented into multiple sub threads, I still don't have the original. But I don't get the impression it was long. Even a rant requires length, but it's more vague than a tirade. :-P -- Chris Murphy -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 On 2015-08-03 23:35, Chris Murphy wrote:
has fragmented into multiple sub threads, I still don't have the original.
Are you using google webmail? Google breaks threading, the OP will be shown by gmail as a different thread. - -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" (Minas Tirith)) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iF4EAREIAAYFAlW/4EMACgkQja8UbcUWM1wbIwD+IHrSbQHMhOb0QND8Pupq/I0T X5d0ucDZOBZEllTxmN0BAJrl8CBglG+Q2fYn4UaMvpwNuX6n+NBqRlFG+ZcdbNC0 =K5Fc -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On Mon, Aug 3, 2015 at 3:42 PM, Carlos E. R. <robin.listas@telefonica.net> wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256
On 2015-08-03 23:35, Chris Murphy wrote:
has fragmented into multiple sub threads, I still don't have the original.
Are you using google webmail? Google breaks threading, the OP will be shown by gmail as a different thread.
Yes but then it also got renamed more than once. IMO, since someone indicated a thread break by changing the subject, it's better that the thread splits. But I have several splits without changed subjects, which might have to do with the embargo, or might have to do with people joining in the discussion who don't get emails (they copy/pasted from archives rather than replying). -- Chris Murphy -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 On 2015-08-04 00:38, Chris Murphy wrote:
On Mon, Aug 3, 2015 at 3:42 PM, Carlos E. R. <> wrote:
Are you using google webmail? Google breaks threading, the OP will be shown by gmail as a different thread.
Yes but then it also got renamed more than once. IMO, since someone indicated a thread break by changing the subject, it's better that the thread splits. But I have several splits without changed subjects, which might have to do with the embargo, or might have to do with people joining in the discussion who don't get emails (they copy/pasted from archives rather than replying).
I only have one break, caused by you replying to an archived mail. All the other emails are properly threaded, with a single parent email :-) It is google email web client which does a lousy job of thread display, although it does a proper job with the headers :-p - -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" (Minas Tirith)) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iF4EAREIAAYFAlW/81QACgkQja8UbcUWM1zAmQD/Rv9XWDSa+gfvxGYTjzuj3wqg MWc7mIYamgPOjLXdS0YBAJO67vyMDOrVrkgJ0SW8qjy+KMzEOHF7BsU88dnM8sRy =NsK9 -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 07/31/2015 09:52 PM, Linda Walsh wrote:
--- the various sendmail rc scripts that did a good job of extracting options from the /etc/sendmail.cf config file -- replaced by "semi-hidden" (in /etc/mail/system -- not a standard location for storing service start/stop scripts -- except w/systemd.
A wonderful benefit -- used to be the scripts were all in /etc/init.d/xxx but now, they can all be put in private, non-standard locations to obfuscate how systemd runs.
There's a point where your vitriol obscure the point you're trying to make. I'm a Postix user, not a sendmail user, but I do not that the start and stop for Postfix are commands that are outside systemd and outside /etc/init.d; quite simply "postfix start" and "postfix stop". Its there very clearly in the man page. Its fundamental to Postfix; should there ever be anything in systemd or init.d to do this then its just a 'wrapper'. It has nothing to do with any benefit or shortcoming of init.d or systemd. A lot of things are like that. Tell me that sendmail that I don't have is a not a regular binary in one of the well known directories, probably /usr/sbin or /sbin, that is is not a regular binary that can be executed from the command line just like Postfix. That even the init.d was not just a shell script wrapper around that binary. There's a "Bah Humbug" -- as Dickens would say -- moment here somewhere. I recall using sendmail on the PDP-11 and early VAXen under BSD back in the 1980s long, long before USG came up with the regularization of the init scripts. BSD start-up back them was a moras of unstructured scripts. The move to the init.d format was welcomed by some as a rationalization and condemned by others as a large business conglomerate forcing its way of doing things onto the creative radical principles that were the founding principles of UNIX. Same old same old comes around again. There's nothing holy about the init.d scripts. It used to be that the UNIX End Of The World was set for January 19th 2038. That supposed we are all still using 32-bit systems. No doubt there will still be some old, old embedded systems that do. But given the rate of innovation in the world its quite likely that even 64-bit machines will have been outdated, and the thread-wars in forums like this will be about the die-hards and dinosaurs who want to keep the now outdated systemd in the face of whatever 'new and innovative' system management and initialization tool a bright, creative but socially inept developer has released though his Avatar. Same old same old comes around again Boring as hell. Same as arguing about KDE17. -- 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 07/31/2015 08:52 PM, Linda Walsh wrote:
A wonderful benefit -- used to be the scripts were all in /etc/init.d/xxx but now, they can all be put in private, non-standard locations to obfuscate how systemd runs.
This is the real crux of the issue. With sysvinit, we all knew to look in /etc/rc.d/rcfoo and to look for yastisms/base configs in /etc/sysconfig. With systemd (which I like, and concur it is just issues such as this that leave a bad taste in my mouth) all scripts/services were to live in /lib/systemd/system with configured services linked below the various /etc/systemd/system locations (e.g. multi-user.target.wants) i.e.: ls -1 /etc/systemd/system/multi-user.target.wants avahi-daemon.service cronie.service cups.path dhcpd4.service dovecot.service httpd.service mysqld.service named.service netctl@rlf_network\x2dstatic.service nmbd.service ntpd.service nut-server.service org.cups.cupsd.path postfix.service remote-fs.target smbd.service spamassassin.service sshd.service vsftpd.service systemd is well behaved and reasonably manageable in that setup (That was with Arch). The additional difficulties suse faces is the extended /etc/sysconfig interface that systemd was loosely shoehorned into and vice versa. That opens up a whole host of new locations where fragmented bits and pieces of configs, services, scripts, etc.. can find themselves stored. When, as in Linda's case, we fail to conform to standard locations or explain changes from the prior logical setup to a new setup/config that seems to have no relation to where standard files are placed regarding systemd, we are only screwing the users out of easy/sane management of their system. This after they have presumably taken the time to learn where and how the duct-tape and bailing-wire of systemd fits together. It is frustrating to take time to learn the proper placement of config/service files for a new system only to find out "somebody took it up themselves move/place/hide a config in a non-standard location anyway". Head scratching?? Why do we have these suggested standards if nobody takes the time to follow them, ... or follow them consistently. systemd works and works well when fully implemented the way Freedesktop.org recommends. (Archlinux is the poster-child for this implementation) openSuSE has several additional challenges in deciding how/where to interface with the sysconfig layout. It would seem that the systemd standard locations for services/etc should control and then via reference or source any information needed from elsewhere. That would at least provide a road-map for the systemd implementation that conforms with the freedesktop recommendations. Having worked though systemd implementation with Arch and now openSuSE, the cleaner and simpler and more consistently you stick to the systemd recommendations, the easier it is for all. -- David C. Rankin, J.D.,P.E. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 08/05/2015 12:44 AM, David C. Rankin wrote:
systemd is well behaved and reasonably manageable in that setup (That was with Arch). The additional difficulties suse faces is the extended /etc/sysconfig interface that systemd was loosely shoehorned into and vice versa. That opens up a whole host of new locations where fragmented bits and pieces of configs, services, scripts, etc.. can find themselves stored.
When, as in Linda's case, we fail to conform to standard locations or explain changes from the prior logical setup to a new setup/config that seems to have no relation to where standard files are placed regarding systemd, we are only screwing the users out of easy/sane management of their system.
This after they have presumably taken the time to learn where and how the duct-tape and bailing-wire of systemd fits together. It is frustrating to take time to learn the proper placement of config/service files for a new system only to find out "somebody took it up themselves move/place/hide a config in a non-standard location anyway". Head scratching?? Why do we have these suggested standards if nobody takes the time to follow them, ... or follow them consistently.
Good question. perhaps the openSuse developers/packagers will answer that. No, wait! what I mean to say is "perhaps we can expect some self-serving justification ..."
systemd works and works well when fully implemented the way Freedesktop.org recommends. (Archlinux is the poster-child for this implementation) openSuSE has several additional challenges in deciding how/where to interface with the sysconfig layout. It would seem that the systemd standard locations for services/etc should control and then via reference or source any information needed from elsewhere. That would at least provide a road-map for the systemd implementation that conforms with the freedesktop recommendations.
Sometimes people obsess with backward compatibility to a point where it is actually crippling. It's when you break away from that you get real innovation. The 'horseless carriage' is a good example. Just the transition from the 'carriage' to the Model-T was a huge watershed. The same thing happened with cell phones. Early phones were in a satchel and had a hand-set like the old "Bakelite' phones. The move to the RIM with display and keyboard was the precursor of a tool that went way beyond anything a 'speaking telegraph' could do.
Having worked though systemd implementation with Arch and now openSuSE, the cleaner and simpler and more consistently you stick to the systemd recommendations, the easier it is for all.
That all says it very well. "I'm a bear of very little brain and long words confuse me" Well long lists and the "sui generis" approach of Windows do. What you rightly point out is that the long standing adage of UNIX/Linux that one learns a few patterns and then applies them (which is what, surely, underpins things like mathematics [and perhaps even arithmetic!] has been broken, and not simply in the way that Market Street "breaks" the street-grid pattern in SanFrancisco, but rather brings in a bucket list of 'exceptions'. Well to be fair, sysvinit was not perfectly 'regular' either... You - almost - make a case for giving up on openSuse and using Arch. Perhaps what we need to do is pressure the openSuse developers to declare a road map for full systemd conformance. Along the way, maybe Linda will announce that while she's giving up on openSuse in frustration she is not a turncoat, not a traitor and is not going to replace openSuse with Windows, but rather with Arch Linux :-0 Perhaps I'll try installing Arch on one of my clunk-servers, the 800Mhz jobs out of the Closet of Anxieties. -- 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 Wed, Aug 5, 2015 at 7:13 AM, Anton Aylward <opensuse@antonaylward.com> wrote:
You - almost - make a case for giving up on openSuse and using Arch.
Or Fedora. :-D
Perhaps what we need to do is pressure the openSuse developers to declare a road map for full systemd conformance.
That's what I think still needs to be done, and made much more concise. I definitely am verbose at times but anytime a development change needs to happen, concise non-rants that embrace consistency and standardization have a much greater chance of getting action. I like the analogy of duct tape and bailing wire, that's developer read worthy too, because who wants to learn non-standard temporary things that are just going to get thrown out eventually anyway and confuse people in the meantime? -- Chris Murphy -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 08/05/2015 08:13 AM, Anton Aylward wrote:
That all says it very well. "I'm a bear of very little brain and long words confuse me" Well long lists and the "sui generis" approach of Windows do. What you rightly point out is that the long standing adage of UNIX/Linux that one learns a few patterns and then applies them (which is what, surely, underpins things like mathematics [and perhaps even arithmetic!] has been broken, and not simply in the way that Market Street "breaks" the street-grid pattern in SanFrancisco, but rather brings in a bucket list of 'exceptions'. Well to be fair, sysvinit was not perfectly 'regular' either...
That is a more than fair statement.
You - almost - make a case for giving up on openSuse and using Arch.
That was not the intent at all. The intent was to show that systemd can be a reasonably friendly environment as long as the freedesktop recommendations for service files and config files are used consistently. The reason this works so flawlessly on Arch, is that is all Arch does. Pure systemd period. No backwards compatible init script hooks, no GUI system configuration tool, nothing, just pure systemd. In that configuration, systemd has no hidden parts, and it is easy to understand. openSuSE has always had, and will most likely always have some yast type GUI configurator tool. That in it self adds a whole new layer of where base configuration tidbits are stored. Instead of simply having the service files control application config and startup, the /etc/sysconfig layout is also incorporated. That's not a bad thing. Yast is what makes openSuSE new user friendly. Rather than having new users dive directly into config files with vi, the base config in /etc/sysconfig can be to a limited extent configured by point-and-click. The challenge is to integrate the systemd service files/scripts with the needed configs held under /etc/sysconfig without mucking up the basic layout and location of each in the process.
Perhaps what we need to do is pressure the openSuse developers to declare a road map for full systemd conformance.
That is the holy grail. If we had a simple rule that said all application provided systemd service and script files remain in their native location under /lib/systemd/system with the machine configuration (the systemd equivalent of checkcfg) comply with the normal scheme of providing links in e.g. /etc/systemd/system/multi-user.target.wants, a clean systemd implementation can be preserved. But how to get the /etc/sysconfig config to dovetail with that systemd layout without creating clutter or scattering files? You simply force the reading/combining to the configs under /etc/systemd/system/ (most easily by sourcing the needed information into whatever application service or script link at that location. You then have the normal locations for the sysconfig and systemd files preserved and a single point of machine config for systemd under /etc/systemd/system.
Along the way, maybe Linda will announce that while she's giving up on openSuse in frustration she is not a turncoat, not a traitor and is not going to replace openSuse with Windows, but rather with Arch Linux :-0
Perhaps I'll try installing Arch on one of my clunk-servers, the 800Mhz jobs out of the Closet of Anxieties.
Use the beginner install wiki. The wiki is excellent. There is NO installer for Arch, so you will be building the install by hand at the console ;-) It is actually a great experience that everyone should go through. There is a great deal of learning that greatly helps with the understanding of the install process. https://wiki.archlinux.org/index.php/Beginners%27_guide -- David C. Rankin, J.D.,P.E. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
participants (9)
-
Andrei Borzenkov -
Anton Aylward -
Carlos E. R. -
Chris Murphy -
David C. Rankin -
David T-G -
greg.freemyer＠gmail.com -
Linda Walsh -
Marcus Meissner