Hi Claudio,
I made a test with NFS mount with nfsvers=4.1 and CentOS 7 as NFS client
(our Netapp already have NFS 4.1 enabled) but the problem is still present.
More, I don't like to switch to NFS 4 because is statefull, NFS v3 is
stateless and for example during maintanace or upgrade of NFS server
clients haven't problems, the reboot of Netapp is trasparent.
I don't think the problem is related to Netapp, I see the same error in
a setup of a customer based on Google Cloud (Ubuntu as Dovecot and NFS
client and Google Cloud NFS volume as storage).
In my case I'm using LDA for local delivery of emails so I hope that
swithcing to LMTP I will resolve the issue but I'm not use since others
users said that they are aready using LMTP.
I don't know why on old Linux distro works and recents distro have the
issue ...
Il 19/01/21 20:21, Claudio Cuqui ha scritto:> It's a long shot......but I would try to use nfsvers=4.1 in the nfs
> mount option (instead of nfsvers=3)? - if your netapp supports it -
> with a newer kernel - 4.14-stable or 4.19-stable (if possible). The
> reason for that, is a nasty bug found in linux nfs client with older
> kernels...
>
>
https://about.gitlab.com/blog/2018/11/14/how-we-spent-two-weeks-hunting-an-nfs-bug/
>
> Hope this helps...
>
> Regards,
>
> Claudio
>
>
> Em qua., 13 de jan. de 2021 ?s 12:18, Maciej Milaszewski
> <maciej.milaszewski at iq.pl <mailto:maciej.milaszewski at
iq.pl>> escreveu:
>
> Hi
> I have been trying resolve my problem with dovecot for a few days
> and I
> dont have idea....
>
> My environment is: dovecot director+5 dovecot guest
>
> dovecot-2.2.36.4 from source
> Linux 3.16.0-11-amd64
> storage via nfs (NetApp)
>
> all works fine but when I update OS from debian 8 (kernel 3.16.x) to
> debian 9 (kernel 4.9.x ) sometimes I get random in logs:
> Broken dovecot-uidlist
>
> examle:
> Error: Broken file
> /vmail2/po/pollygraf.xxx_pg_pollygraf/Maildir/dovecot-uidlist line 88:
> Invalid data:
>
> (for random users - sometimes 10 error in day per node, some times
> more)
>
> File looks ok
>
> But if I change kernel to 3.16.x problem with "Broken file
> dovecot-uidlist"? - not exists
> if turn to 4.9 or 5.x - problem exists
>
> I have storage via nfs with opions:
>
rw,sec=sys,noexec,noatime,tcp,hard,rsize=65536,wsize=65536,intr,nordirplus,nfsvers=3,tcp,actimeo=120
> I tested with "nocto" or without "nocto" - nothing
changes ......
>
> nfs options in node:
> mmap_disable = yes
> mail_fsync = always
>
> I bet the configuration is correct and I wonder why the problem occurs
> with other kernels
> 3.x.x - ok
> 4.x - not ok
>
> I check and user who have problem did not connect to another node in
> this time
>
> I dont have idea why problem exists on the kernel 4.x but not in 3.x
>
>
--
Alessio Cecchi
Postmaster @ http://www.qboxmail.it
https://www.linkedin.com/in/alessice
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<https://dovecot.org/pipermail/dovecot/attachments/20210122/f104aff3/attachment-0001.html>