Since one hour I got each 60 seconds interval those messages:
Aug 22 23:52:03 antares syslog-ng[1091]: Error suspend timeout has elapsed, attempting to write again; fd='11'
Aug 22 23:52:03 antares syslog-ng[1091]: syslog-ng internal() messages are looping back, preventing loop by suppressing further messages; recurse_count='2'
Aug 22 23:52:03 antares syslog-ng[1091]: Suspending write operation because of an I/O error; fd='11', time_reopen='60'
Aug 22 23:53:03 antares syslog-ng[1091]: Error suspend timeout has elapsed, attempting to write again; fd='11'
Aug 22 23:53:03 antares syslog-ng[1091]: I/O error occurred while writing; fd='11', error='Input/output error (5)'
Aug 22 23:53:03 antares syslog-ng[1091]: Suspending write operation because of an I/O error; fd='11', time_reopen='60'
Aug 22 23:54:03 antares syslog-ng[1091]: Error suspend timeout has elapsed, attempting to write again; fd='11'
Aug 22 23:54:03 antares syslog-ng[1091]: I/O error occurred while writing; fd='11', error='Input/output error (5)'
Aug 22 23:54:03 antares syslog-ng[1091]: Suspending write operation because of an I/O error; fd='11', time_reopen='60'
Aug 22 23:55:03 antares syslog-ng[1091]: Error suspend timeout has elapsed, attempting to write again; fd='11'
Aug 22 23:55:03 antares syslog-ng[1091]: I/O error occurred while writing; fd='11', error='Input/output error (5)'
Aug 22 23:55:03 antares syslog-ng[1091]: Suspending write operation because of an I/O error; fd='11', time_reopen='60'
Aug 22 23:56:03 antares syslog-ng[1091]: Error suspend timeout has elapsed, attempting to write again; fd='11'
Aug 22 23:56:03 antares syslog-ng[1091]: I/O error occurred while writing; fd='11', error='Input/output error (5)'
Aug 22 23:56:03 antares syslog-ng[1091]: Suspending write operation because of an I/O error; fd='11', time_reopen='60'
I downloaded the syslog-ng 3.2 guide but I really don’t understand what it is all about… and, then, how interpret the error/warning (?) messages.
I did an hibernate before leaving but this is not the first time I put my computer in hibernation and I have never had such messages after the recovery of the pages snapshots…
I know what is a I/O error but I don’t know which device is involved (hard disk ? device memory ? …)
I will first try a reboot and then go back to indicate if the issue disappeared… after reboot still no more such messages…
Am 23.08.2012 00:16, schrieb soundlord:
> Aug 22 23:56:03 antares syslog-ng[1091]: Suspending write operation because of an I/O error; fd=‘11’, time_reopen=‘60’
The file descriptor seems to be 11, so do a
readlink /proc/self/fd/11
to get the filename.
–
PC: oS 12.2 x86_64 | i7-2600@3.40GHz | 16GB | KDE 4.8.4 | GeForce GT 420
ThinkPad E320: oS 12.1 x86_64 | i3@2.30GHz | 8GB | KDE 4.8.5 | HD 3000
eCAFE 800: oS 12.1 i586 | AMD Geode LX 800@500MHz | 512MB | KDE 3.5.10
Am 23.08.2012 00:28, schrieb Martin Helm:
> Am 23.08.2012 00:16, schrieb soundlord:
>> Aug 22 23:56:03 antares syslog-ng[1091]: Suspending write operation because of an I/O error; fd=‘11’, time_reopen=‘60’
>
> The file descriptor seems to be 11, so do a
>
> readlink /proc/self/fd/11
>
to get the filename.
Sorry that was stupid, I am too tired, since you see no process id you
can check for all process id’s which have a fd=11 with something like
that (I am sure someone else can do better it is hot and late here)
for x in $(echo /proc/*/fd/11); do echo $x $(readlink $x); done
The command lsof should also be helpful in such case.
–
PC: oS 12.2 x86_64 | i7-2600@3.40GHz | 16GB | KDE 4.8.4 | GeForce GT 420
ThinkPad E320: oS 12.1 x86_64 | i3@2.30GHz | 8GB | KDE 4.8.5 | HD 3000
eCAFE 800: oS 12.1 i586 | AMD Geode LX 800@500MHz | 512MB | KDE 3.5.10
Thank you martin ^^
This is the process linked with the pid : /proc/1072/fd/11 /dev/tty10
In fact the number between square brackets “” “]” is the pid.
It happens only when my computer wakes from hibernation.
Am 28.08.2012 03:16, schrieb soundlord:
>
> Thank you martin ^^
>
> This is the process linked with the pid : /proc/1072/fd/11 /dev/tty10
> In fact the number between square brackets “” “]” is the pid.
>
> It happens only when my computer wakes from hibernation.
>
>
Sounds as if it is syslog itself which has a problem when trying to
write to the virtual console on tty10.
I have no real idea how that can be related to the wakeup.
You could check if after wakeup messages are still written to tty10 by
watching it with [CTRL][ALT][F10].
–
PC: oS 12.2 x86_64 | i7-2600@3.40GHz | 16GB | KDE 4.8.4 | GeForce GT 420
ThinkPad E320: oS 12.1 x86_64 | i3@2.30GHz | 8GB | KDE 4.8.5 | HD 3000
eCAFE 800: oS 12.1 i586 | AMD Geode LX 800@500MHz | 512MB | KDE 3.5.10
I will try, till now I just killed the process attached to tty10… then there were no more messages about this I/O error.
I’m reading the documentation about syslog-ng and to be honnest it is ununderstandable… I don’t know why suddenly (since I updated the kernel) I have such issue.
I will dig…
All I can see is that the rights are not the same for all the tty devices in /dev
885 0 crw--w---- 1 root root 4, 0 Aug 30 17:46 tty0
890 0 crw-rw---- 1 root tty 4, 1 Aug 30 17:47 tty1
899 0 crw-r----- 1 root root 4, 10 Aug 30 17:47 tty10
891 0 crw-rw---- 1 root tty 4, 2 Aug 30 17:47 tty2
909 0 crw--w---- 1 root tty 4, 20 Aug 30 17:46 tty20
892 0 crw-rw---- 1 root tty 4, 3 Aug 30 17:47 tty3
919 0 crw--w---- 1 root tty 4, 30 Aug 30 17:46 tty30
893 0 crw-rw---- 1 root tty 4, 4 Aug 30 17:47 tty4
929 0 crw--w---- 1 root tty 4, 40 Aug 30 17:46 tty40
894 0 crw-rw---- 1 root tty 4, 5 Aug 30 17:47 tty5
939 0 crw--w---- 1 root tty 4, 50 Aug 30 17:46 tty50
895 0 crw-rw---- 1 root tty 4, 6 Aug 30 17:47 tty6
949 0 crw--w---- 1 root tty 4, 60 Aug 30 17:46 tty60
896 0 crw--w---- 1 root tty 4, 7 Aug 30 17:46 tty7
897 0 crw--w---- 1 root tty 4, 8 Aug 30 17:46 tty8
898 0 crw--w---- 1 root tty 4, 9 Aug 30 17:46 tty9
Why tty0 and tty10 are not set in the tty group ?
Is there a way to be sure that all the tty devices are set to tty group ?
…and I just remember, the tty 7 to 12 are reserved for the X windows system (for X sessions).
tty 11 and 12 have the same permissions than tty7 to tty9…
The only difference is the group that is not the same and (of course) the write access for group member that is not set for tty10.
Nov 14 20:05:26 antares syslog-ng[1097]: I/O error occurred while writing; fd='11', error='Input/output error (5)'
Nov 14 20:05:26 antares syslog-ng[1097]: Suspending write operation because of an I/O error; fd='11', time_reopen='60'
Nov 14 20:06:26 antares syslog-ng[1097]: Error suspend timeout has elapsed, attempting to write again; fd='11'
Nov 14 20:06:26 antares syslog-ng[1097]: I/O error occurred while writing; fd='11', error='Input/output error (5)'
Nov 14 20:06:26 antares syslog-ng[1097]: Suspending write operation because of an I/O error; fd='11', time_reopen='60'
…every minute a new sequence of those type of messages are added to the log… each time I do a hibernate.
for x in $(echo /proc/*/fd/11); do echo $x $(readlink $x); done
/proc/10007/fd/11 socket:[29779]
/proc/10015/fd/11 socket:[29769]
/proc/1097/fd/11 /dev/tty10
/proc/1115/fd/11 socket:[10413]
/proc/1275/fd/11 socket:[6177]
/proc/1405/fd/11 /dev/ati/card0
/proc/2054/fd/11 socket:[6581]
/proc/2279/fd/11 anon_inode:inotify
/proc/2369/fd/11 socket:[7258]
/proc/3040/fd/11 socket:[8108]
/proc/3045/fd/11 /var/log/mysql/antares_queries.log
/proc/3048/fd/11 socket:[8308]
/proc/3087/fd/11 /var/lib/samba/sessionid.tdb
/proc/3094/fd/11 /var/lib/samba/sessionid.tdb
/proc/3100/fd/11 anon_inode:[eventfd]
/proc/3435/fd/11 /dev/tty0
/proc/3510/fd/11 pipe:[29872]
/proc/3650/fd/11 socket:[11092]
/proc/3664/fd/11 socket:[10412]
/proc/3678/fd/11 socket:[10482]
/proc/3757/fd/11 /dev/network_latency
/proc/3798/fd/11 /proc/mdstat
/proc/3802/fd/11 socket:[11609]
/proc/3865/fd/11 anon_inode:inotify
/proc/3871/fd/11 anon_inode:inotify
/proc/3872/fd/11 anon_inode:inotify
/proc/3991/fd/11 pipe:[12335]
/proc/3993/fd/11 pipe:[12396]
/proc/4006/fd/11 socket:[13126]
/proc/4040/fd/11 pipe:[12952]
/proc/4135/fd/11 socket:[13241]
/proc/4136/fd/11 socket:[13235]
/proc/4137/fd/11 socket:[13284]
/proc/4138/fd/11 socket:[13248]
/proc/4180/fd/11 socket:[13479]
/proc/4191/fd/11 socket:[13508]
/proc/4391/fd/11 pipe:[14308]
/proc/4397/fd/11 pipe:[14361]
/proc/7057/fd/11 anon_inode:inotify
/proc/9977/fd/11 pipe:[29545]
It is seemly the tty that is the cause of this I/O issue…
This is clearly a permissions issue but I can’t change the permissions on this tty10 as it will not cause the log to have added lines every minutes untill I kill it…
It is said that the tty are set up at boot with their default permissions (tty10 got different permissions compared with the standard ones tty0 to tty12).
Could somebody has an idea how to proceed to set the default permissions back to tty10 ?
Thanx