NetXMS Support Forum

We have a lot of events "SYS_THREAD_HANG" (Thread "Item Poller" is not responding).
Most of time only for a few seconds, then "SYS_THREAD_RUNNING" again.

What is the best way to analyse this and find the root cause?
(maybe a good point for the troubleshooting paragraph of the administrator guide)

In server log we get:
[14-Apr-2015 15:30:51.580] [ERROR] Thread "Item Poller" does not respond to watchdog thread

Server debug log is attached.

We use NetXMS V2.0-M3 (linux x64, PostgreSQL).

Thanks!

Ok thanks Alex for the explanation. So if I need a MAC address human readable I have to manually convert it in an external converter.

Especially with MAC addresses we see situations in SNMP values like this:

ARP-Table:
.1.3.6.1.2.1.3.1.1.2.8.1.192.168.30.3 [Hex-STRING] = 4C 9E FF 6F E0 17
.1.3.6.1.2.1.3.1.1.2.8.1.192.168.30.4 [Hex-STRING] = 4C 9E FF 6F E2 15
.1.3.6.1.2.1.3.1.1.2.8.1.192.168.30.10 [Hex-STRING] = 00 08 9B DA 8B C5
.1.3.6.1.2.1.3.1.1.2.8.1.192.168.30.100 [STRING] = x~a?G
.1.3.6.1.2.1.3.1.1.2.8.1.192.168.30.103 [STRING] = Tä:?ön
.1.3.6.1.2.1.3.1.1.2.8.1.192.168.30.109 [Hex-STRING] = 28 6A BA 90 1B 77
.1.3.6.1.2.1.3.1.1.2.8.1.192.168.30.111 [STRING] = ô·âw_

Some values are of type STRING instead of Hex-STRING (in the same SNMP table!). We see this especially with ZyXEL devices, but other vendors too. Is this a problem of the monitored devices or in NetXMS? (NetXMS is V2.0-M3 on Linux x64, built from sources). If you need more information, no problem. Thanks.

agent config (put together from server agent configuration and agent policy):

MasterServers = netxms.domain.com
ListenPort = 4700
RequireAuthentication = yes
RequireEncryption = yes
EnabledCiphers = 1
SharedSecret = longkey
EnableWatchdog = yes
LogFile = {syslog}
LogFile = ${ProgramFiles}\NetXMS\nxagentd.log
DebugLevel = 0
ConfigIncludeDir = ${ProgramFiles}\NetXMS\etc\nxagentd.conf.d
FileStore = ${ProgramFiles}\NetXMS\var
SubAgent = ecs.nsm
SubAgent = portcheck.nsm
SubAgent = ping.nsm
SubAgent = winperf.nsm
SubAgent = wmi.nsm
SubAgent = logwatch.nsm
SubAgent = filemgr.nsm

[filemgr]
RootFolder = ${SystemDrive}\

[logwatch]
ProcessOfflineEvents = yes
Parser = ${ProgramFiles}\NetXMS\var\eventlog-system-parser.xml
Parser = ${ProgramFiles}\NetXMS\var\eventlog-application-parser.xml

Registry Key HKEY_USERS\.DEFAULT\Software\NetXMS\LogParserLibrary:
LogWatch.Application REG_DWORD 0x552b7a44
LogWatch.System REG_DWORD 0x552b7a78

We also see /0 netmasks instead of /24 on HP printers since upgrade from M2 to M3 (server on Linux x64 built from sources).

Hi Victor. Thanks for the quick fix. PING subagent is working again! Thanks, Dani

For which version of NetXMS is this scheduled now? (capturing Windows Critical level events) Or should I register a feature request?

Hi Victor

I come back to #2 and #4:

#2: I tested again with V2.0.-M3 but it is not working for me. Can you please check again?

#4: For which version is this planned?

Thanks, Dani

Victor

To the "PING subagent" issue:
I made a network dump and saw the IP addresses the subagent uses are in reverse order:
set in agent config: 172.16.50.1
used from PING subagent: 1.50.16.172

I hope you have a short solution, I can fix in the source code.

Thanks Victor, I can confirm that fixed the one with Wake-on-LAN (for managed and unmanaged nodes).

We use IP addresses in PING subagent config:
[ping]
DefaultPacketSize = 46
PacketRate = 4
Target = 172.16.50.1:host.doamin.intern


PING subagent:
Since upgrade from V2.0-M2 to V2.0-M3 I have a problem with PING subagent, with no change in the existing configuration I get now always Latency=10k (=timeout) (Icmp.AvgPingTime).
We use NetXMS on Linux x64 compiled from sources. PING subagent is on the NetXMS server.
Is there any known problem?

VPN-connector Subnets tab:
Since V2.0-M3 I can't open the Subnets tab of an existing VPN-connector anymore, I get the following error:
"The currently displayed page contains invalid values." in the Windows x64 Java Console.

Wake-on-LAN:
Now Wake-on-LAN is not working anymore (managed or unmanaged node).

sorry, it was not the right feed. I was a little bit too fast.
Try to switch 'SyslogNodeMatchingPolicy' to '1' in Server Configuration and restart the NetXMS server. That changes source to node matching.
0 = "syslog message source IP address then hostname "
1 = "hostname then syslog message source IP address"

maybe check this:
https://www.netxms.org/forum/general-support/syslog-from-source-unknown/15/

edit:
sorry wrong link

thanks for that!