Package: ntp Version: 1:4.2.6.p5+dfsg-2 Severity: important X-Debbugs-CC: debian-ad...@lists.debian.org User: debian-ad...@lists.debian.org Usertags: needed-by-DSA-Team
DSA recently setup a couple of machines with a rootfs on NFS
(mips-aql-01/2). Both of them sometimes start getting errors like these
from the ntp cron job. Once the errors start, they do not stop until the
ntp service is restarted. Looking at the cron job I think the fix would
be to replace the find command with this one. It would be great if you
could get this included in the next Debian point release.
find "$statsdir" -name '*stats*' -type f -mtime +7 -exec rm {} \;
/etc/cron.daily/ntp:
rm: cannot remove `/var/log/ntpstats/.nfs00000000032a208e0005399d': Device or
resource busy
pabs@mips-aql-01:~$ sudo lsof | grep -F .nfs | grep ntp
ntpd 7496 ntp 6w REG 0,14 58
53092494 /var/log/ntpstats/.nfs00000000032a208e0005399d
(172.29.172.4:/raid0/data/_NAS_NFS_Exports_/root-mips-aql-01)
pabs@mips-aql-01:~$ sudo service ntp restart
[ ok ] Stopping NTP server: ntpd.
[ ok ] Starting NTP server: ntpd.
pabs@mips-aql-01:~$ sudo lsof | grep -F .nfs | grep ntp
pabs@mips-aql-01:~$ ls /var/log/ntpstats/.nfs00000000032a208e0005399d
ls: cannot access /var/log/ntpstats/.nfs00000000032a208e0005399d: No such file
or directory
pabs@mips-aql-01:~$ sudo lsof | grep ntpstats
ntpd 21100 ntp 4w REG 0,14 44467
53089010 /var/log/ntpstats/peerstats.20141211
(172.29.172.4:/raid0/data/_NAS_NFS_Exports_/root-mips-aql-01)
ntpd 21100 ntp 5w REG 0,14 159645
53085160 /var/log/ntpstats/cryptostats.20141211
(172.29.172.4:/raid0/data/_NAS_NFS_Exports_/root-mips-aql-01)
ntpd 21100 ntp 6w REG 0,14 58
53087774 /var/log/ntpstats/loopstats.20141211
(172.29.172.4:/raid0/data/_NAS_NFS_Exports_/root-mips-aql-01)
--
bye,
pabs
https://wiki.debian.org/PaulWise
signature.asc
Description: This is a digitally signed message part
