Geoff,
>
> Over the weekend we received these error messages from an mv5500 running
> vxworks 5.5 and epics 3.14.6. As far as I can tell this is an ENOBUFS
> error. Is this the correct interpretation?
Yes, that is the ENOBUFS error number. The default error number to string
converter uses strerror(), and on your build of vxWorks strerror() returns
the string "errno = 0x37".
> What buffer needs to overflow to get this error?
The ENOBUFS error number is a catch all for any pool exhaustion problems in
the IP kernel. With newer versions of vxWorks the size of the IP stack's
buffer pool is fixed at compile time. See the debugging section of the CA
reference manual.
> I thought the only UDP
> messages from an IOC were beacons to port 5065. What are the other UDP
> messages to ports other than 5065? Could this indicate a problem
> somewhere else on our network?
The CA server is sending a search response message to address
"131.225.231.247:35139". Presumably you have a CA client whose ephemeral UDP
port assignment on 131.225.231.247 is 35139.
> Could this indicate a problem somewhere else on our network?
If your network stack pool has a standard out-of-the-box configuration then
the most likely cause is local (pool allocation configuration that does not
match up with the load).
You might also consider running casw in order to track you beacon anomaly
rates. Beacon anomalies can be used to identify IOCs with weak network
connections.
Jeff
> -----Original Message-----
> From: Geoff Savage [mailto:[email protected]]
> Sent: Tuesday, July 05, 2005 8:29 AM
> To: EPICS Tech-Talk
> Subject: UDP send, errno=0x37
>
> Hi,
>
> Over the weekend we received these error messages from an mv5500 running
> vxworks 5.5 and epics 3.14.6. As far as I can tell this is an ENOBUFS
> error. Is this the correct interpretation?
>
> CAS: UDP send to "131.225.231.247:35139" failed because "errno = 0x37"
> CAS: UDP send to "131.225.231.247:35139" failed because "errno = 0x37"
> ../online_notify.c: CA beacon (send to "131.225.231.255:5065") error was
> "errno
> = 0x37"
> CAS: UDP send to "131.225.231.247:35139" failed because "errno = 0x37"
> CAS: UDP send to "131.225.231.65:33655" failed because "errno = 0x37"
> CAS: UDP send to "131.225.231.65:33655" failed because "errno = 0x37"
> CAS: UDP send to "131.225.231.65:33655" failed because "errno = 0x37"
> CAS: UDP send to "131.225.231.247:35139" failed because "errno = 0x37"
> CAS: UDP send to "131.225.231.247:35139" failed because "errno = 0x37"
> CAS: UDP send to "131.225.231.247:35139" failed because "errno = 0x37"
> CAS: UDP send to "131.225.231.65:33655" failed because "errno = 0x37"
> CAS: UDP send to "131.225.231.65:33655" failed because "errno = 0x37"
> CAS: UDP send to "131.225.231.65:33655" failed because "errno = 0x37"
> CAS: UDP send to "131.225.231.247:35139" failed because "errno = 0x37"
>
> What buffer needs to overflow to get this error? I thought the only UDP
> messages from an IOC were beacons to port 5065. What are the other UDP
> messages to ports other than 5065? Could this indicate a problem
> somewhere else on our network?
>
> Thanks, Geoff
- References:
- UDP send, errno=0x37 Geoff Savage
- Navigate by Date:
- Prev:
RE: UDP send, errno=0x37 Allison, Stephanie
- Next:
RE: odd libca message printed Jeff Hill
- Index:
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
<2005>
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
- Navigate by Thread:
- Prev:
UDP send, errno=0x37 Geoff Savage
- Next:
RE: UDP send, errno=0x37 Allison, Stephanie
- Index:
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
<2005>
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
|