[opensuse] snapper
Hello, I see in the snapper pages https://en.opensuse.org/Portal:Snapper it's said than snapper can do high system i/o "Now you might also want to disable cleanup of empty pre-post snapshot pairs since otherwise the daily cleanup can cause high I/O" * may this look like a system crash (disabling even the mouse for a long time - several minutes?) * is it possible to instruct snapper to do less often snapshots? I see nothing like this in yast I guess that the very often seen lines below are this adctivity logs? déc. 07 16:00:01 linux-uegt CRON[4120]: pam_unix(crond:session): session closed for user root déc. 07 16:00:01 linux-uegt systemd[4122]: pam_unix(systemd-user:session): session closed for user root déc. 07 16:15:01 linux-uegt cron[4201]: pam_unix(crond:session): session opened for user root by (uid=0) déc. 07 16:15:01 linux-uegt systemd[4202]: pam_unix(systemd-user:session): session opened for user root by (uid=0) déc. 07 16:15:01 linux-uegt CRON[4201]: pam_unix(crond:session): session closed for user root déc. 07 16:15:01 linux-uegt systemd[4203]: pam_unix(systemd-user:session): session closed for user root thanks jdd -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 12/07/2014 10:53 AM, jdd wrote:
Hello,
I see in the snapper pages
https://en.opensuse.org/Portal:Snapper
it's said than snapper can do high system i/o
"Now you might also want to disable cleanup of empty pre-post snapshot pairs since otherwise the daily cleanup can cause high I/O"
The "daily" and the "cron" should be a clue. Or you could RTFM of SNAPPER(1) and read <quote> Automatic Snapshot Creation Next to manual snapshot creation, snapshots are also created automatically. · A cron-job creates hourly snapshots. · Certain programs like YaST and zypper create pre/post snapshot pairs when modifying the system. </quote> Which might encourage you to look in /etc/cron.{daily,hourly} and find the files "suse-de.sanpper"
* may this look like a system crash (disabling even the mouse for a long time - several minutes?)
* is it possible to instruct snapper to do less often snapshots? I see nothing like this in yast
No, its in the config file for snapper. Stop treating yast like a know-it-all and a crutch and investigate the FS. RTFM time again: <quote> FILES /etc/sysconfig/snapper Global configuration file. /etc/snapper/configs Directory containing configuration files. </quote> -- /"\ \ / ASCII Ribbon Campaign X Against HTML Mail / \ -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Le 07/12/2014 17:11, Anton Aylward a écrit :
On 12/07/2014 10:53 AM, jdd wrote:
Hello,
I see in the snapper pages
https://en.opensuse.org/Portal:Snapper
it's said than snapper can do high system i/o
"Now you might also want to disable cleanup of empty pre-post snapshot pairs since otherwise the daily cleanup can cause high I/O"
The "daily" and the "cron" should be a clue.
not really they do not say if this could block a computer
Or you could RTFM of SNAPPER(1) and read
Which might encourage you to look in /etc/cron.{daily,hourly} and find the files "suse-de.sanpper"
but in these files all is "no"
* may this look like a system crash (disabling even the mouse for a long time - several minutes?)
no anwser to this, but as if I understand there are no snapshots (apart for running yast), it shouldn't
Stop treating yast like a know-it-all and a crutch and investigate the FS.
I just do what the wiki says to do...
RTFM time again:
there are no hourly nor daily snapshots there so the main answer is no, the logs are not from snapper... jdd -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 12/07/2014 12:27 PM, jdd wrote:
there are no hourly nor daily snapshots there
My read of the cron functions and the man page is that they do clean-up, so the absence of snapshots tells me they are working properly :-) what does # snapper list tell you?
so the main answer is no, the logs are not from snapper...
There should be /var/log/snapper.log to cross check. Is there that file? What time are the entries? If they are nothing to do with snapper then why did you bring it up? If you really are concerned with the cron records its easy enough to add a few lines to the /usr/lib/cron/run-crons script to tell what is being done. Oh, and what other cron jobs do you have configured? They may account for this. -- /"\ \ / ASCII Ribbon Campaign X Against HTML Mail / \ -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Le 07/12/2014 18:41, Anton Aylward a écrit :
On 12/07/2014 12:27 PM, jdd wrote:
there are no hourly nor daily snapshots there
My read of the cron functions and the man page is that they do clean-up, so the absence of snapshots tells me they are working properly :-)
no, they are simply disabled everywhere, so cron have nothing to do :-) )
what does
# snapper list
tell you?
only yast and zypp (but I use it a lot at the moment)
Oh, and what other cron jobs do you have configured? They may account for this.
none but I found the answer in crontab, I guess (there are so many place where a cronjob can be defined, it's easy to forget one :-) thanks jdd -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 12/07/2014 12:55 PM, jdd wrote:
no, they are simply disabled everywhere, so cron have nothing to do
Perhaps that's why the logs you show us show cron starting, doing nothing and so reporting nothing, and exiting. -- /"\ \ / ASCII Ribbon Campaign X Against HTML Mail / \ -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 12/07/2014 12:55 PM, jdd wrote:
what does
# snapper list
tell you?
only yast and zypp (but I use it a lot at the moment)
Sorry, I obviously wasn't explicit enough. Could you please show us what the output of that command is. -- /"\ \ / ASCII Ribbon Campaign X Against HTML Mail / \ -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 12/07/2014 12:55 PM, jdd wrote:
Oh, and what other cron jobs do you have configured? They may account for this.
none
but I found the answer in crontab, I guess (there are so many place where a cronjob can be defined, it's easy to forget one :-)
And what was the answer then? -- /"\ \ / ASCII Ribbon Campaign X Against HTML Mail / \ -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Le 07/12/2014 18:59, Anton Aylward a écrit :
On 12/07/2014 12:55 PM, jdd wrote:
what does
# snapper list
tell you?
only yast and zypp (but I use it a lot at the moment)
Sorry, I obviously wasn't explicit enough.
Could you please show us what the output of that command is.
not very informative ype | # | Pre # | Date | Utilisateur | Nettoyer | Description | Données utilisateur -------+-----+-------+--------------------------------+-------------+----------+-------------------+-------------------- single | 0 | | | root | | current | pre | 128 | | jeu. 04 déc. 2014 22:11:24 CET | root | number | zypp(packagekitd) | important=no post | 129 | 128 | jeu. 04 déc. 2014 22:12:04 CET | root | number | | important=no pre | 134 | | ven. 05 déc. 2014 11:50:36 CET | root | number | yast printer | post | 135 | 134 | ven. 05 déc. 2014 12:02:27 CET | root | number | | pre | 136 | | sam. 06 déc. 2014 09:54:38 CET | root | number | zypp(packagekitd) | important=yes post | 137 | 136 | sam. 06 déc. 2014 09:55:15 CET | root | number | | important=yes pre | 138 | | dim. 07 déc. 2014 16:25:48 CET | root | number | yast firewall | post | 139 | 138 | dim. 07 déc. 2014 16:26:21 CET | root | number | | pre | 140 | | dim. 07 déc. 2014 16:39:06 CET | root | number | yast snapper | post | 141 | 140 | dim. 07 déc. 2014 16:45:04 CET | root | number | | -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Le 07/12/2014 19:00, Anton Aylward a écrit :
On 12/07/2014 12:55 PM, jdd wrote:
Oh, and what other cron jobs do you have configured? They may account for this.
none
but I found the answer in crontab, I guess (there are so many place where a cronjob can be defined, it's easy to forget one :-)
And what was the answer then?
I openned an other thread for it in /etc/crontab, I read: # check scripts in cron.hourly, cron.daily, cron.weekly, and cron.monthly # -*/15 * * * * root test -x /usr/lib/cron/run-crons && /usr/lib/cron/run-crons >/dev/null 2>&1 the question is why these logs come in now and never did before? but it's not snapper related. I have frequent crashes and I try to understand the logs to found why. I learn a lot, but nothing relevant to the crash until now :-( jdd -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 12/07/2014 01:01 PM, jdd wrote:
Le 07/12/2014 18:59, Anton Aylward a écrit :
On 12/07/2014 12:55 PM, jdd wrote:
what does
# snapper list
tell you?
only yast and zypp (but I use it a lot at the moment)
Sorry, I obviously wasn't explicit enough.
Could you please show us what the output of that command is.
not very informative
ype | # | Pre # | Date | Utilisateur | Nettoyer | Description | Données utilisateur -------+-----+-------+--------------------------------+-------------+----------+-------------------+--------------------
single | 0 | | | root | | current | pre | 128 | | jeu. 04 déc. 2014 22:11:24 CET | root | number | zypp(packagekitd) | important=no post | 129 | 128 | jeu. 04 déc. 2014 22:12:04 CET | root | number | | important=no pre | 134 | | ven. 05 déc. 2014 11:50:36 CET | root | number | yast printer | post | 135 | 134 | ven. 05 déc. 2014 12:02:27 CET | root | number | | pre | 136 | | sam. 06 déc. 2014 09:54:38 CET | root | number | zypp(packagekitd) | important=yes post | 137 | 136 | sam. 06 déc. 2014 09:55:15 CET | root | number | | important=yes pre | 138 | | dim. 07 déc. 2014 16:25:48 CET | root | number | yast firewall | post | 139 | 138 | dim. 07 déc. 2014 16:26:21 CET | root | number | | pre | 140 | | dim. 07 déc. 2014 16:39:06 CET | root | number | yast snapper | post | 141 | 140 | dim. 07 déc. 2014 16:45:04 CET | root | number | |
It is if you know how to read it. For example: where did the first 127 go? Did you delete them manually or were they deleted 'automatically' by a cron job at the turnover of the month? Why was one packagekit update important and another not? I get the impression that you you don't drill down and investigate very much; so many of you questions are ones that a reading of the manual pages or the helo files or simply tracing through 'causality' would resolve. No need to be Sherlock Holmes, just methodological. -- /"\ \ / ASCII Ribbon Campaign X Against HTML Mail / \ -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 12/07/2014 01:04 PM, jdd wrote:
I openned an other thread for it
in /etc/crontab, I read:
# check scripts in cron.hourly, cron.daily, cron.weekly, and cron.monthly # -*/15 * * * * root test -x /usr/lib/cron/run-crons && /usr/lib/cron/run-crons >/dev/null 2>&1
the question is why these logs come in now and never did before?
but it's not snapper related.
No, this is the same tread. You obviously weren't paying attention. That is the script that runs the /etc/cron.{hourly,daily}/* scripts, which I've already mentioned, in particular the suse.de-snapper. So yes it is snapper related. Its also related to making backups, going out to the network for spamassassin updates, and other things, any of which might, if configured the right way, cause your system to hang while they are working, even if they don't error. Do note, as we've discussed here before, that the cron jobs are actually run 'niced' down in priority and which records the output of the job. While I'm being methodological about it, I'll also pint out that there is the shell variable SYSLOG_ON_NO_ERROR. Unless that is set to "yes" then you are not going to see any details about the cron.{hourly,daily} jobs unless they error. So the absence of details on your log reports other than the privilege escalation makes perfect sense. There's also the samell matter of who the result gets sent to. How this works depends on how you have your system setup. If, for example, you read your mail via Thunderbird or Kmail then make sure its set up to read local mail in /var/spool/mail. And make sure that you have Postscript or Exim or Sendmail doing the aliasing properly so system mail, mail to root, gets sent to an account your ARE monitoring. This might have been set up when you did your install, but do check that. Be methodical. -- /"\ \ / ASCII Ribbon Campaign X Against HTML Mail / \ -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Le 07/12/2014 19:42, Anton Aylward a écrit :
On 12/07/2014 01:01 PM, jdd wrote:
not very informative
It is if you know how to read it.
I guess I know. I see twice a day, no hourly (as configured), yast conrresponf to me using yast and zypp runs approx once a day
For example: where did the first 127 go? Did you delete them manually or were they deleted 'automatically' by a cron job at the turnover of the month?
I was seeking for job running every 15 minutes, i guess I found them
I get the impression that you you don't drill down and investigate very much;
I spent approx 3 hours today on this, don't think it's small (too much in my idea) so many of you questions are ones that a reading of the manual
pages or the helo files or simply tracing through 'causality' would resolve. No need to be Sherlock Holmes, just methodological.
I am, so why I don't continue on closed question... thanks jdd -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Le 07/12/2014 19:54, Anton Aylward a écrit :
On 12/07/2014 01:04 PM, jdd wrote:
I openned an other thread for it
in /etc/crontab, I read:
# check scripts in cron.hourly, cron.daily, cron.weekly, and cron.monthly # -*/15 * * * * root test -x /usr/lib/cron/run-crons && /usr/lib/cron/run-crons >/dev/null 2>&1
the question is why these logs come in now and never did before?
but it's not snapper related.
No, this is the same tread. You obviously weren't paying attention.
please stop being rude, it's of no use
That is the script that runs the /etc/cron.{hourly,daily}/* scripts, which I've already mentioned, in particular the suse.de-snapper.
but these scripts runs conditional that all ends up doing nothing
So yes it is snapper related.
see the answer http://lists.opensuse.org/opensuse/2014-12/msg00362.html "I also saw these annoying logs, but didn't find a working idea how to avoid or suppress this"
Do note, as we've discussed here before, that the cron jobs are actually run 'niced' down in priority and which records the output of the job.
but there is only one cron job running (the one quoted) every 15 minutes. This is a pretty new install, with nothing cronned by me
While I'm being methodological about it, I'll also pint out that there is the shell variable SYSLOG_ON_NO_ERROR. Unless that is set to "yes" then you are not going to see any details about the cron.{hourly,daily} jobs unless they error.
this is a good idea, done
do check that. Be methodical.
I am, it's because I am that I think I can exclude snapper, having verifyed all the files (thanks to your help) I read root mails. The only one is from mdadm script: "ls: cannot access /dev/md?*: No such file or directory" normal, given I have no lvm. thanks jdd -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
participants (2)
-
Anton Aylward -
jdd