Bug 110060

Summary:	ucd-snmp 4.2.3 - net-snmp 5.0.6 still hangs snmpd, system, life
Product:	[Retired] Red Hat Linux	Reporter:	Ben Lentz <blentz>
Component:	net-snmp	Assignee:	Phil Knirsch <pknirsch>
Status:	CLOSED DUPLICATE	QA Contact:	David Lawrence <dkl>
Severity:	high	Docs Contact:
Priority:	medium
Version:	9	CC:	rvokal
Target Milestone:	---
Target Release:	---
Hardware:	i386
OS:	Linux
Whiteboard:
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2006-02-21 19:00:00 UTC	Type:	---
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:

Description Ben Lentz 2003-11-14 16:24:23 UTC

Description of problem:
This is probably a duplicate of 82146. I tried to add a comment here, 
but Bugzilla told me I was trying to change the Component field from 
AfterStep-APPS to net-snmp and that I wasn't privileged enough. Hmm.
My apologies for the DUPL.

Not to rehash and old problem, but I'm having the exact same problem. 
I've got a base Red Hat Linux 9 system with net-snmp-5.0.6. I can 
start the daemon, but snmpwalk (or an external snmp walking program) 
either returns partial output and then kills the listener (Timeout: 
No Response from localhost) or consumes the CPU ~95%, with the logs 
(snmpd -D -l /var/log/snmpd.log ..etc) showing an infinate loop when 
viewed with tail -f. 

It appears that this bug has existed since RHL7.2, in v4.2.3. Is this 
going to be fixed? Is there any way to use this service if it 
repeatedly crashes? The tool I'm using to query SNMP data needs to be 
able to walk the tree to discover the device. 

If there is a place in /etc/snmp/snmpd.conf to change 'system' 
to '.1', I'll do it (the host is RO and behind a firewall) except 
that I can't pinpoint where this change is to be made. Any further 
assistance you can provide would be greatly appreciated.
Thanks for your time.

Version-Release number of selected component (if applicable):
net-snmp-5.0.6-17

How reproducible:
Everytime

Steps to Reproduce:
1.Install RH (7.2-9.0) with the corresponding package containing 
snmpd.
2. Run '/etc/rc.d/init.d/snmpd start'
3. Run 'snmpwalk -v 1 localhost -c public system
  
Actual results:
top shows the snmpd process at ~95%. System no longer responds to 
snmp queries.

Expected results:
snmp should run and respond and not kill the box.

Additional info:
I have seen this bug many places spanning many versions. I'm having a 
tough time grasping that this (seemingly basic) problem has gone 
uncorrected for so long.

Comment 1 Phil Knirsch 2003-11-14 16:52:52 UTC

Hi!

As described in the other bug reports, this is a know problem with the
upstream code and is not likely to be fixed in the near future.

In order to make it at least responsive again look for the 

view    systemview    included

lines and append the following line:

view    systemview    included   .1

That should prevent the snmpd running amok.

Btw: the system recovers after a few minutes, then snmpd goes back to
idle state and can receive new requests. :-)

Read ya, Phil



*** This bug has been marked as a duplicate of 82146 ***

Comment 2 Ben Lentz 2003-11-14 21:09:46 UTC

view    systemview    included   .1

Worked like a charm. You da man, thanks so much for clearing that up 
for me.

Have a good one!

Comment 3 Red Hat Bugzilla 2006-02-21 19:00:00 UTC

Changed to 'CLOSED' state since 'RESOLVED' has been deprecated.