ORACLE WebDB Listener error

[ConnieK]

This may not be the correct forum to ask this question, but my DBA cannot find any reference to this on ORACLE's MetaLink, so... here goes...

I have an HP9000 T520 with HPUX 11.0 running an application that had an ORACLE DB (ver 8.1.6) and an ORACLE WebDB (ver 2.2.0.0.8) running. From time to time, the Web Listener (on port 80) goes down. The listener log shows these two errors: "Attempting to Listen - Fatal Error - Poll failed (errno 0)" and "Attempting to Listen - Fatal Error - Accept failed (errno 233)".
I have had to stop and start the listener process to get it up listening again.

Does anyone have a clue as to why these error messages are generated? Could it be a kernel parameter that needs to be adjusted? A needed Patch? A network problem?

Thanks in advance for your assistance!

Connie

[Sanjay_6]

Hi Constance,

Seems like your system is running out of Buffer Space. The error no 233 refers to ENOBUFS -> No Buffer Space Available

quote //

1. ) No buffer space is available. The accept() cannot complete. The queued socket connect request is aborted.

2.) No buffer space is available. The bind() cannot complete.

3.)No buffer space available. An operation on a socket was not performed because the system lacked sufficient buffer space.

// Endquote

Hope this helps.

Thanks

[Christian Schulze]

I still have this Problem. My workaround is: WebDB-Lsitener is configures as a Service in the MC/Serviceguard environment, number of restarts is indefinite, check Interval is 5 sec.

Christian

[ConnieK]

I've assigned points to the two answers - Thanks to Sanjay & Christian!
I have NOT yet solved this problem, but at least Sanjay gave me a heading. I have asked the DBA to get guidance from our Oracle Support Office. I believe the problem is related to some of the kernel parameters and I assume that somewhere there is a list of recommended kernel parm settings for this configuration of the WebDB.
Thanks Again.

[george anderson_7]

Constance, If you find a solution to this problem. I would love to hear about it. I have been fighting against this problem with the webdb listener for the past 10 months. please drop me an email to geoande1@yahoo.com


thanks

[Printaporn_1]

Hi,

I also donno for the right solution , but my WebDB process is running very fine , but what diffrent is I use port 22005 and following is some of My Kernel ( I am using HP-UX 11.0 64bits)

bufcache_hash_locks 128
bufpages (NPROC*3)
dbc_max_pct 10
dbc_min_pct 2
maxdsiz 0X40000000
maxdsiz_64bit 0x0000000040000000
maxfiles 2048
maxfiles_lim 2048
maxqueuetime 0
maxssiz 0x00800000
maxssiz_64bit 0x00800000
maxswapchunks 4096
maxtsiz 0x04000000
maxtsiz_64bit 0x0000000040000000
maxuprc ((NPROC*9)/10)
maxusers 200
nbuf 0
nproc ((MAXUSERS*3)+64)
maxusers 200

[Tim Krego_2]

We had similar problems with our WebDB listener about a year ago. We are using an Oracle 8.1.6 database, WebDB 2.2.0.0.5 running on a K460 HPUX 11.00 box.

Here is a script we use to monitor the WebDB listener and restart if necessary. Change the email address if you want to receive notifications on your pager/cellphone.