master1 and master2 are BIND servers, providing IXFR.
All was good yesterday, after “nsdc patch” NSD reported about ~7000000 records processed normally.
Today I got:
zonec: processed 621379 RRs in “example.com”.
Did somebody meet the same problem? Any ideas how to solve and not meet this problem again?
Were you out of disk-space (so, the file was truncated) ? Did the
computer reset during the nsd-patch zone write or something like that?
No for both questions.
Otherwise, do you have the last changes (ixfr.db, nsd.db, … other
unnatural incidents) for debug analysis?
I made a lot of “update-rebuild-patch” trying to retransfer full zone, so these files are changed heavily.
Another unnatural incident is that when I do “nsdc patch”, NSD says “warning: slave zone example.com with no zonefile ‘/srvs/nsd/etc/slave/example.com.S’(No such file or directory) will force zone transfer.” but no file appearing, neither right after command issued, nor after some time passed.
ls -l /srvs/nsd/etc
total 16
drwxr-xr-x 2 apn dns 512 Mar 27 11:59 RCS
-r–r–r-- 1 apn dns 7479 Mar 27 11:59 nsd.conf
drwxrwxr-x 2 bind dns 512 Mar 29 14:52 slave
master1 and master2 are BIND servers, providing IXFR.
All was good yesterday, after "nsdc patch" NSD reported about ~7000000 records processed normally.
Today I got:
zonec: processed 621379 RRs in "example.com".
Did somebody meet the same problem?
Yes we did , but we were unable to reproduce the problem.
Any ideas how to solve and not meet this problem again?
NSD 3.2.9 introduced a flaw in zone transfer, which reduces the speed of the transfer greatly. I would recommend upgrading to 3.2.10, like Vicky suggested.
NSD 3.2.9 introduced a flaw in zone transfer, which reduces the speed of the transfer greatly. I would recommend upgrading to 3.2.10, like Vicky suggested.
Best regards,
Matthijs
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hello list,
I encountered weird problem with NSD 3.2.9. Suddenly it lost about 90% of zone. The configuration is following:
master1 and master2 are BIND servers, providing IXFR.
All was good yesterday, after “nsdc patch” NSD reported about ~7000000 records processed normally.
Today I got:
zonec: processed 621379 RRs in “example.com”.
Did somebody meet the same problem?
Yes we did , but we were unable to reproduce the problem.
Any ideas how to solve and not meet this problem again?