SMS 8240 Disk is full.

[Migration User]

Hi

I have a SMS 8240 that reports, "Minimum queue disk bytes free limit reached, queue inbound free 454MB limit 500MB" 
Is there a way that i can change the limit of the queue, as a short time solution?

Yesterday we startet having problems with mail taking long time to gett inn and out from us, afther some checking i fond that the queue had reached its 500MB free diskspace limit.
In our appliance there are 37 GB disks, in the Local Host hardware Status, Disk Usage is shown in red and says 22.6 GB.

Any Suggestions?


Regards
CT

[Migration User]

[Migration User]

Increasing the queue size limit will make your situation worse, not better. If you totally run out of disk you'll be stuck. You need to figure out what is causing the delivery delays, and using up your disk. Is you 8240 the only box? - if not, are other scanners also out of disk space? Is it your control center? - What's your log retention? Can you afford to shorten the retention time and purge the database? - Are you keeping backups on-box? How many are you retaining? Can you get rid of some Queues: - what queue is filling - inbound, outbound or delivery? - Is there a domain stuck? To get a count of message by domain use mta-control all active-routes - SBG 9 only: what is the common error? from SUPPORT command line, use mta-control delivery query-queue | grep "Error" | sort to see what delivery errors you are seeing. (you can pipe this to a file if you want to look at it with less) mta-control delivery query-queue | grep "Error" | sort > /tmp/queue-errors.txt less /tmp/queue-errors.txt Disk usage Login as support, if you are on SBG 9, what does the output of monitor disk command report? disk: data_used data_free opt_used opt_free other_used other_free 8360836 109116620 358832 1567084 781840 6922036 or SBG 8 use du -h Filesystem Size Used Avail Use% Mounted on /dev/sda5 7.8G 764M 6.7G 11% / /dev/sda3 99M 12M 83M 13% /boot /dev/sda8 119G 8.0G 105G 8% /data /dev/sda7 2.0G 351M 1.5G 19% /opt none 2.0G 0 2.0G 0% /dev/shm

[Migration User]

Where in contact with Symantec support, who found out that the disk was indeed full, we ran the "delete all command" which helped a bit, but still data disk was 83% full, we/he then navigated trough the data disk and deleted some other files, I don't quite recall what files it was but it made a big difference in disk usage.
Have made some changes to the log retention, and will monitor the disk usage closely for a while to see if something unusual happens.

To your questions:

We have one 8240
The log retention was 20 days, have changed it to 7.
The queues was not filling up, but the disk was to full (something eat up the disk space), so the appliance halted in and outgoing mail.
There has to be a minimum of 500MB of free disk space in order to send/receive mail.

CT

[Migration User]

Logging level and retention can mess things up.   We keep log type = warning, except for Mail Transfer Agent = Information. 
Database storage limits are 50 MB and 7 day. Message audit log is enabled and set for 30 days (I think Symantec recommends only 7 days).

The reporting database can really eat disk space as well.   We track sender domains, sender IP, recipients domains.  We are on SBG 9.01 so we also track recipients and invalid recipients, retention is 45 days (Sym recommends 30 days).

We are on 8380's  830 GB - 4 scanners, 1 dedicated CC, 6 M messages/day.

We do have some 8260's running SBG 9.01.  In some versions of SBG, the remote logging didn't work very well, but syslog feature now appears stable, and all the message audit stuff gets written to the syslog box.  You could implement that and off load logs if you need to keep more than the 7 days Symantec recommended.   SysLog Watcher, Personal Edition will work for testing (only allows a single syslog source).