Core files were found: smbd.core - SMD service down

TrueNAS General
,
1

This happens the second time now (on several TN Core machines):

New alerts:
* The following system core files were found: smbd.core. Please create a ticket
at httXX://ixsystems.atlassian.net/ and attach the relevant core files along
with a system debug. Once the core files have been archived and attached to
the ticket, they may be removed by running the following command in shell: ‘rm
/var/db/system/cores/*’.

I can see that the /var volume is small

Filesystem Size Used Avail Capacity Mounted on
boot-pool/ROOT/13.0-U6.8 5.5G 1.3G 4.2G 24% /
devfs 1.0K 1.0K 0B 100% /dev
tmpfs 32M 9.5M 22M 30% /etc
tmpfs 4.0M 8.0K 4.0M 0% /mnt
tmpfs 2.6G 2.6G 0B 100% /var
fdescfs 1.0K 1.0K 0B 100% /dev/fd

and full, although I cannot find where the data is that filled it up.

I have no idea why smbd has crashed. A reboot helped last time but the problem came back roughly after 1 week.

I am seeing a lot of No space left on device and NT_STATUS_DISK_FULL in log.smbd

The pool volumes are far from 100% full.

Any ideas?

2

So if you run 100% out of space on a device, then smbd (and potentially many other processes) will crash. I’ve seen this before when running out of storage entirely on data pool, but in your case it may be a process writing to one of your tmpfs filesystem. I vaguely recall there were some edge cases where cloud sync failure logs could build up in the rundir and go OOM.

3

But the logs are not on tmpfs, they are mounted to ZFS.

4

There are various samba state files that are on tmpfs.