Failed to Semaphore with Linux agent

[bugjargal]

Hi all,

I'm meeting the same issue as joekro has last summer in thread478-1265730. The only difference being that I'm running RHEL3 instead of Centos.

Does any body have a clue about what could be going wrong.

Thanks in advance.

 

[amacquet]

Hello,

I'm facing the same problem, I get this in uag.log :

08/10 22:53:24(3958) - W300014: (GetIPC) Failed to get semaphore (No space left on device)
08/10 22:53:24(3958) - (node_backup) Failed in GetIPC

Does anyone have a solution ?

thx !

Fernando

 

[davidmichel]

Got this from a friend who is up on UNIX.

Basically this means that the process could not get enough Shared Ram. On Windows Shared ram is created as needed. On Unix (except for AIX) the admin sets the amount of ram aside. Basically the client needs to increase the # of semaphore’s on his system, it may require a reboot.

 

[amacquet]

Hi,

thank you for replying. Actually there were many zombie processes on the machine, and many semaphores blocked by them.

After a reboot everything went fine again.

regards,
Antoine