Hi All,
We have an issue with a esxi host on 6.0 where the tmp directly is full.
The error receive: "The ramdist 'tmp' is full. As a result, the file /tmp/altbootbank.ick could not be written."
Ran a vdf -h.
if we run an 'ls -al' there are no files that would use up that space. Even after checking hidden directories.
There are plenty of inodes available.
Any ideas how to free up this space?
Thanks!
Hi,
What ever is saved in /tmp directory is non persistent, means if the host reboot all the contents inside it will be deleted.
So you can go ahead and delete the contents which you think can bee deleted from your end.
-Sachin
Hello DarkSaber,
Can you share the output of running this on the host? (and or use it to check for anything anomalous or large yourself)
#find /tmp/ -exec ls -larth '{}' \;
Bob
Do not see anything larger than 4M:
~ # find /tmp/ -exec ls -larth '{}' \;
total 4080
-rw-r--r-- 1 root root 0 Oct 25 02:06 mili2d.log
-rw------- 1 root root 0 Oct 25 20:10 1xcib3
-rw------- 1 root root 0 Oct 25 20:13 .snmpd-started
-rw------- 1 root root 0 Oct 25 20:14 lzKS7S
-rw------- 1 root root 0 Oct 25 20:27 TuQ9DA
-rw------- 1 root root 0 Oct 28 15:25 D7WgTe
-rw-r--r-- 1 root root 0 Oct 28 15:27 snmpd-vm-report.xml
-rw------- 1 root root 0 Oct 28 15:28 glO3F5
-rw------- 1 root root 0 Oct 28 19:51 ctoloR
-rw------- 1 root root 0 Oct 28 20:09 RDpkzw
drwx------ 1 root root 512 Oct 28 20:09 vmware-root
-rw------- 1 root root 0 Oct 28 20:09 wPY7uD
drwxr-xr-x 1 root root 512 Nov 2 19:23 ..
-rw------- 1 root root 0 Nov 2 19:30 cdce6g
-rw-r--r-- 1 root root 4.0M Nov 2 19:56 ql_ima_sdm.log
-rw-r--r-- 1 root root 0 Nov 2 19:56 ql_ima.log
-rw-r--r-- 1 root root 0 Nov 2 20:01 altbootbank.lck
-rw------- 1 root root 0 Nov 2 20:12 ojKM68
-rw------- 1 root root 36 Nov 2 20:35 probe.session
-rw-r--r-- 1 root root 0 Nov 2 20:35 snmpd-cvtcimsnmp.xml
drwxrwxrwt 1 root root 512 Nov 2 20:35 .
-rw------- 1 root root 36 Nov 2 20:35 /tmp/probe.session
-rw-r--r-- 1 root root 4.0M Nov 2 19:56 /tmp/ql_ima_sdm.log
total 12
drwxr-xr-x 1 root root 512 Oct 28 20:09 ha-agentmgr
drwx------ 1 root root 512 Oct 28 20:09 .
drwxrwxrwt 1 root root 512 Nov 2 20:35 ..
total 8
-rw------- 1 root root 0 Oct 28 20:09 upgrade.sig
-rw------- 1 root root 0 Oct 28 20:09 upgrade
drwx------ 1 root root 512 Oct 28 20:09 ..
drwxr-xr-x 1 root root 512 Oct 28 20:09 .
-rw------- 1 root root 0 Oct 28 20:09 /tmp/vmware-root/ha-agentmgr/upgrade
-rw------- 1 root root 0 Oct 28 20:09 /tmp/vmware-root/ha-agentmgr/upgrade.sig
-rw-r--r-- 1 root root 0 Oct 25 02:06 /tmp/mili2d.log
-rw-r--r-- 1 root root 0 Nov 2 19:56 /tmp/ql_ima.log
-rw------- 1 root root 0 Oct 25 20:10 /tmp/1xcib3
-rw------- 1 root root 0 Oct 25 20:13 /tmp/.snmpd-started
-rw------- 1 root root 0 Oct 25 20:14 /tmp/lzKS7S
-rw------- 1 root root 0 Oct 25 20:27 /tmp/TuQ9DA
-rw------- 1 root root 0 Oct 28 15:25 /tmp/D7WgTe
-rw-r--r-- 1 root root 0 Oct 28 15:27 /tmp/snmpd-vm-report.xml
-rw------- 1 root root 0 Oct 28 15:28 /tmp/glO3F5
-rw------- 1 root root 0 Oct 28 19:51 /tmp/ctoloR
-rw------- 1 root root 0 Oct 28 20:09 /tmp/RDpkzw
-rw------- 1 root root 0 Oct 28 20:09 /tmp/wPY7uD
-rw------- 1 root root 0 Nov 2 19:30 /tmp/cdce6g
-rw-r--r-- 1 root root 0 Nov 2 20:01 /tmp/altbootbank.lck
-rw------- 1 root root 0 Nov 2 20:12 /tmp/ojKM68
-rw-r--r-- 1 root root 0 Nov 2 20:35 /tmp/snmpd-cvtcimsnmp.xml
Just a correction;
vcenter is 6.0
The host itself is on 5.5.
Hi,
What ever is saved in /tmp directory is non persistent, means if the host reboot all the contents inside it will be deleted.
So you can go ahead and delete the contents which you think can bee deleted from your end.
-Sachin
Thanks, we ended up rebooted the host and the problem was resolved.
Interestingly it ended up being that the coredumps were being directed there.
We had to go through this technote to redirect the coredumps: VMware Knowledge Base
Not sure why they didn't show up...
Thanks for the help!