James,
Thanks for you help!
The event window freezes exactly the same time as the nvserverd log, and if I close it or restart the nvserverd daemon nothing happens. I have to make a full stop/restart daemons for the system to return to full operation.
There are no queues shown in netstat -a.
About the ESE.automation, only the valid ruleset is listed in the file.
nvtecia -stop and -reload did not restore the event flow, although I did receive a message in nvserverd regarding the adapter restart:
07/30/2007 08:53:39 AM Close of TEC API requested.
07/30/2007 08:53:39 AM TEC_ITS_BASE;source=nvserverd;origin=10.96.1.13;hostname=xxxxxxxx.com.br
;adapter_host=xxxxxxxx.com.br;date="07/30/2007 08:53:39 AM";status=OPEN;severity=FATAL;msg="nvserverd is ending";END
07/30/2007 08:53:39 AM Shutdown Event sent
07/30/2007 08:53:39 AM nvcorrd session closed
07/30/2007 08:53:39 AM TEC handle destroyed
07/30/2007 08:53:39 AM Close of TEC API completed.
07/30/2007 08:54:15 AM Restart of TEC API requested.
07/30/2007 08:54:15 AM Reading
tecint.conf parameters:
07/30/2007 08:54:15 AM NvserverdTraceTecEvents = YES
07/30/2007 08:54:15 AM fallbackClass = TEC_ITS_BASE
07/30/2007 08:54:15 AM ServerLocation = @EventServer
07/30/2007 08:54:15 AM Java Options = -Xusealtsigs
07/30/2007 08:54:15 AM NvserverdPrimeTecEvents = (null)
07/30/2007 08:54:15 AM NvserverdSendSeverityTecEvents = (null)
07/30/2007 08:54:15 AM TecRuleName = zacchi.rs
07/30/2007 08:54:15 AM TecConnServer::openSession( ) with nvcorrd:
07/30/2007 08:54:15 AM fRuleName = zacchi.rs, CorrelationType = 1, priming_flag = 1, TivoliTEC = 1
07/30/2007 08:54:15 AM TEC_ITS_BASE;source=nvserverd;origin=10.96.1.13;hostname=
xxxxxxxx.com.br;adapter_host=xxxxxxxx.com.br;date="07/30/2007 08:54:15 AM";status=CLOSED;severity=HARMLESS;msg="nvserverd has connected using rule zacchi.rs";END
Beyond that, nothing else is logged on the file and no events are forwarded to TEC.
I am already in contact with IBM support but, still, if anyone has any idea, I am open minded.
Best regards,
Marcelo
On 7/27/07, James Shanks
<jshanks@us.ibm.com> wrote:
Please try not to make as fool of me. One cannot make a remote diagnosis if one does not have all the facts. The hung event window was an important piece of information which you left out the first time. But it remains true that once an event is given to the TEC interface nvserverd has no knowledge of it.
This forum is not a substitute for Level 2, I can only advise you on what steps I might take to try to narrow down the problem. But I have not seen this hang issue here.
Is the event window frozen at the same place as the nvserverd.log? If you close it, do events start flowing to TEC? If you do a netstat -a do you see send and receive queues backed up? If you open a new event window is that also hung? Are there other rulesets running in the background via
ESE.automation? Have you tried using nvtecia -stop and stopping the TEC flow to see whether the events windows start flowing again?
Perhaps there is a daemon problem, though none of our SuSe Linux systems show such an issue here and the fact that it is isolated to just one machine makes it even less likely that it is a code bug. If all else fails, then the only thing I can think of is to obtain (1) a debug nvserverd from NetView L2, and then (2) run a diagnostic tool they can also give you called gdbprof. If nvserverd is hung somewhere, then that will show conclusively that it is. But it will not tell you what to do about it. That may take more diagnosis and trial-and-error.
Hope this helps. Best I can do.
James Shanks
Level 3 Support for Tivoli NetView for UNIX and Windows
Network Availability Management
Network Management - Development
Tivoli Software, IBM Corp
![Inactive hide details for "Marcelo Zacchi" <mzacchi@gmail.com>]() "Marcelo Zacchi" <
mzacchi@gmail.com>
James,
The fact that the "Current Events" window also freezes also does not indicate a problem with the daemon?
On 7/27/07, James Shanks <
jshanks@us.ibm.com> wrote:
All nvserverd does is format the message and give it to the TEC adapter (EEIF) library functions. The code provided by TEC does the sending. Once nvserverd passes a formatted message along to TEC, he is ignorant of what happens to it. If you see it logged in the nvserverd log, and not followed by an error message, then the hand-off to the TEC library went smoothly. If it works for a while and then stops, that is a TEC issue and not a NetView one.
You will need to get a TEC EEIF trace and let TEC support tell you what is happening. NetView has no way of knowing. NetView L2 can help you enlist the aid of TEC support but that is all.
James Shanks
Level 3 Support for Tivoli NetView for UNIX and Windows
Network Availability Management
Network Management - Development
Tivoli Software, IBM Corp
![Inactive hide details for "Marcelo Zacchi" <mzacchi@gmail.com>]()
"Marcelo Zacchi" <
mzacchi@gmail.com>
James and Praveen,
I am having the exact same problem on my Linux box.
I have disable the state correlation but it didn't work. I have been exchanging traces and logs with the IBM support but, until now, no conclusion has been reached.
The funniest thing is that it appears to 'last' for approximately 12 hours and then it simply stops. The nvserverd.log only brings messages of successful operations, no error messages whatsoever.
I have installed FixPack 1 too, but that did not solve the problem.
I am running NV on a SUSE LINUX Enterprise Server 9 (x86_64), but I have a environment that is just like this one and does not have any problem with nvserverd.
Has anyone been able to solve or workaround this?
Best regards to all,
Marcelo Zacchi
Siemens SIS NOC LATAM
On 12/13/06, James Shanks <
jshanks@us.ibm.com > wrote:
Have you tried anything to isolate the problem? Is your TEC cache file growing? Then the TEC code is having problems talking to your TEC library.
Are events still being written to the nvserverd.log? If you don't have one then modify your tecint.conf file by uncommenting the line which says #NvserverdTraceTecEvents=YES.
If the nvserverd.log shows that events are not being sent to TEC, then you can try turning off State Correlation in tecint.conf. Make it say UseStateCorrelation=NO. If that solves the problem then most likely you have a java issue.
James Shanks
Level 3 Support for Tivoli NetView for UNIX and Windows
Network Availability Management
Network Management - Development
Tivoli Software, IBM Corp
![Inactive hide details for "Praveen Kumar" <Praveenkumar.nair@ustri.com>]()
"Praveen Kumar" <
Praveenkumar.nair@ustri.com>
Hi All,
I am running NetView 7.1.5 (upgraded from 7.1.4) on Linux, I have the events forwarded to TEC. After the upgrade nvserverd seems to getting halted (but the status shows running) no events are forwarded until a restart of the nvserverd daemon.
Would anyone please help me?
True regards,
Praveen Kumar._______________________________________________
NV-L mailing list
NV-L@lists.ca.ibm.com
Unsubscribe:NV-L-leave@lists.ca.ibm.com
http://lists.ca.ibm.com/mailman/listinfo/nv-l
(Browser access limited to internal IBM'ers only)
_______________________________________________
NV-L mailing list
NV-L@lists.ca.ibm.com
Unsubscribe:NV-L-leave@lists.ca.ibm.com
http://lists.ca.ibm.com/mailman/listinfo/nv-l
(Browser access limited to internal IBM'ers only)
_______________________________________________
NV-L mailing list
NV-L@lists.ca.ibm.com
Unsubscribe:NV-L-leave@lists.ca.ibm.com
http://lists.ca.ibm.com/mailman/listinfo/nv-l
(Browser access limited to internal IBM'ers only)
_______________________________________________
NV-L mailing list
NV-L@lists.ca.ibm.com
Unsubscribe:NV-L-leave@lists.ca.ibm.com
http://lists.ca.ibm.com/mailman/listinfo/nv-l
(Browser access limited to internal IBM'ers only)
_______________________________________________
NV-L mailing list
NV-L@lists.ca.ibm.com
Unsubscribe:NV-L-leave@lists.ca.ibm.com
http://lists.ca.ibm.com/mailman/listinfo/nv-l (Browser access limited to internal IBM'ers only)
_______________________________________________
NV-L mailing list
NV-L@lists.ca.ibm.com
Unsubscribe:NV-L-leave@lists.ca.ibm.com
http://lists.ca.ibm.com/mailman/listinfo/nv-l
(Browser access limited to internal IBM'ers only)
_______________________________________________
NV-L mailing list
NV-L@lists.ca.ibm.com
Unsubscribe:NV-L-leave@lists.ca.ibm.com
http://lists.ca.ibm.com/mailman/listinfo/nv-l (Browser access limited to
internal IBM'ers only)
|