Hipcrostino
Cadet
- Joined
- May 24, 2012
- Messages
- 2
Hi all,
I've got a FreeNas server setup to do two things. On the one hand its a normal file server in a windows network using CIFS. On the other hand its an ISCSI target for two Windows 2008 R2 Hyper-V machines in a cluster.
Most of the time it works just fine. The ISCSI system is set up on its own separate network through its own switch so that ISCSI traffic isn't clogging the main network. This seems to be working just fine.
My problem is that every couple of weeks the the ISTGT server fails. It gives this error:
"istgt[3006]: istgt_iscsi.c:5088:worker: ***ERROR*** pthread_create() failed" (3006 is the PID an always is)
This is followed by many items saying "last message repeated xx times" where xx can be any number. As a result, the Hyper-V Cluster falls over (just test and dev servers, no big deal) and then soon after CIFS stops responding. The box still responds via keyboard so I tell it to restart from the interface (option 10). When I do this the following happens:
Waiting for PIDS: 3006
Followed 30 seconds later with the watchdog timeout. Shutdown is then aborted, the discs get synced (thank god) and the box stops there. I have to use the reset button to fire it back up.
I've tried to kill the 3006 service manually but no commands seem to do anything. I'd be happy if I could just get the server to restart properly after an error like this.
Server is running 8.0.4 at the moment. I plan to upgrade to p2 ASAP, although there is no information that changes have been made to ISCSI in that time.
Has anyone else got this problem? Its starting to get annoying and I can't move production machines into the Hyper-V cluster until its resolved.
ISTGT version is: "istgt version 0.4 (20110928)"
I've got a FreeNas server setup to do two things. On the one hand its a normal file server in a windows network using CIFS. On the other hand its an ISCSI target for two Windows 2008 R2 Hyper-V machines in a cluster.
Most of the time it works just fine. The ISCSI system is set up on its own separate network through its own switch so that ISCSI traffic isn't clogging the main network. This seems to be working just fine.
My problem is that every couple of weeks the the ISTGT server fails. It gives this error:
"istgt[3006]: istgt_iscsi.c:5088:worker: ***ERROR*** pthread_create() failed" (3006 is the PID an always is)
This is followed by many items saying "last message repeated xx times" where xx can be any number. As a result, the Hyper-V Cluster falls over (just test and dev servers, no big deal) and then soon after CIFS stops responding. The box still responds via keyboard so I tell it to restart from the interface (option 10). When I do this the following happens:
Waiting for PIDS: 3006
Followed 30 seconds later with the watchdog timeout. Shutdown is then aborted, the discs get synced (thank god) and the box stops there. I have to use the reset button to fire it back up.
I've tried to kill the 3006 service manually but no commands seem to do anything. I'd be happy if I could just get the server to restart properly after an error like this.
Server is running 8.0.4 at the moment. I plan to upgrade to p2 ASAP, although there is no information that changes have been made to ISCSI in that time.
Has anyone else got this problem? Its starting to get annoying and I can't move production machines into the Hyper-V cluster until its resolved.
ISTGT version is: "istgt version 0.4 (20110928)"