vmware ESX cause root full

VC cannot manage one ESX Host but the rest can manage


Please note that the ESX host which is 3.5 Base build was seen as not responding in the VC which is of 2.5 U2 due to the "/" partition becoming full.

The "/" partition was filling up due the "/var/core" directory getting full which would happen due to the crashes of the processes inside the Service Console of the ESX.

When I checked the "/var/log/vmware/hostd.log", I found that the agent was back tracing as follows:

=============================================================================================================================

[2010-02-06 17:44:25.062 'DirectorysvcPlugin' 3076449184 info] Plugin initialized
[2010-02-06 17:44:25.062 'App' 3076449184 info] Trying hostsvc
[2010-02-06 17:44:25.070 'HostsvcPlugin' 3076449184 info] Storage data synchronization policy set to invalidate_change
[2010-02-06 17:44:25.070 'Vmfs2UpgradeModeLock' 3076449184 info] [Init] undoAttempts(15), timeoutMsec(2000)
[2010-02-06 17:44:25.419 'HostsvcPlugin' 3076449184 info] Resource pool configuration: /etc/vmware/hostd/pools.xml
[2010-02-06 17:44:25.422 'App' 3076449184 panic] error: FileIO error: No space left : /etc/vmware/hostd/pools.xml.tmp
[2010-02-06 17:44:25.423 'App' 3076449184 panic] backtrace:




Solutions:

1.Log on to the ESX host via putty session, log in as Root
2. Type df -h : To know the free space
3. Go into the directory by type cd /var/core
4. Type ls to list what is inside the core directory
5. Type rm -rf *.core : To delete all the core files. Please note * is used to inform the file names before the .core extension


If root files not full, try the following
service mgmt-vmware stop
service mgmt-vmware start

or

service mgmt-vmware restart