"Disk space below threshold" for NFS-mounted store with 748G available ?

[DCh]

I've got /opt/zimbra/store mounted to an NFSv4 store

zimbra@mail: df -h | grep -i /store
   nas.dchil.loc:/servers/zimbra/store
     788G  197M  748G   1% /opt/zimbra/store

Plenty of space available.

But if I restart zimbra, I get a WARNING about not enough disk space,

WARNING: Disk space below threshold for /opt/zimbra/store.

zimbra@mail-:~$ zmcontrol restart
Host mail.dchil.loc
        Stopping stats...Done.
        Stopping mta...Done.
        Stopping spell...Done.
        Stopping snmp...Done.
        Stopping cbpolicyd...Done.
        Stopping archiving...Done.
        Stopping antivirus...Done.
        Stopping antispam...Done.
        Stopping imapproxy...Done.
        Stopping memcached...Done.
        Stopping mailbox...Done.
        Stopping logger...Done.
        Stopping zmconfigd...Done.
        Stopping ldap...Done.
Host mail.dchil.loc
        Starting ldap...Done.
        Starting zmconfigd...Done.
        Starting logger...Done.
df: no file systems processed
        WARNING: Disk space below threshold for /opt/zimbra/store.
        Starting mailbox...Done.
        Starting antispam...Done.
        Starting antivirus...Done.
        Starting snmp...Done.
        Starting spell...Done.
        Starting mta...Done.
        Starting stats...Done.

Clearly there's enough space.

Is there some Zimbra config I need to do? Or is this a bug?

[Krishopper]

NFS is greatly discouraged. Per Performance Tuning Guidelines for Large Deployments - Zimbra :: Wiki :

NO NFS. It is our experience that the world is full of poor NFS implementations (server and client), and sometimes the disks backing that NFS mount are not performant to boot. Also note that many upstream OSS components of Zimbra (BDB, OpenLDAP, MySQL, Lucene) have or do discourage the use of NFS to store binary/mmaped data.

For remote storage, I'd suggest something like iSCSI so that at least Zimbra can utilize the underlying ext3 filesystem like it expects.

[DCh]

Yes, I've read that page.

It hardly addresses what 'poor' means. Historically, that's been file-locking issues ... also the article comments about "binary/mmaped data". Iiuc that's NOT the case with the mailstore data, is it?

In any case, thanks for the comment -- but that wasn't my question here.

I'm asking about why Zimbra seems to think it doesn't have enough space.

[Krishopper]

Sorry, re: the "not enough space" message -- It might have an issue doing a stat() on an NFS file system. I've never really looked into that code, but it may have something to do that. Since they don't support storing data on NFS, they probably haven't tested it to ensure that it returns the values they expect.