Community
- Command Central
- MVP Program
- Monthly Mission
- Blogs
- Groups
- Events
- Media Vault
Products
- Observability
- Network Management
- Application Management
- IT Security
- IT Service Management
- System Management
- Database Management
Content Exchange
- SolarWinds Platform
- Server & Application Monitor
- Database Performance Analyzer
- Server Configuration Monitor
- Network Performance Monitor
- Network Configuration Manager
- SQL Sentry
- Web Help Desk
Free Tools & Trials
Store

Real-Time Process Explorer: "Value was either too large or too small for a UInt64."

When attempting to run a "Real-Time Process Explorer" on a CentOS 6.5 Linux node, getting:

An exception occurred during polling processes. Error:
Value was either too large or too small for a UInt64.

Screen Shot 2016-12-20 at 1.48.48 PM.png

Process monitors otherwise seem to run fine on it.

Total processes:

$ ps -A --no-headers | wc -l
887
$ ps -AL --no-headers | wc -l
3008

Orion Platform 2016.1.5300, NPM 12.0, SAM 6.2.4.

The node has 192GB memory. SNMP - native to that OS, version 5.5.

Is this a large memory + SNMP related issue?

Not a lot of related threads on Thwack:

APM 1.0 SP4 fixed the issue in 2008
A similar issue with SAM 5.0 RC2 in 2012, fixed in RC3

Find more posts tagged with

Linux

Accepted answers

All comments

Petr.Vilem

Can you find some additional info about this issue in following log when the issue is reproduced?

ProgramData\Solarwinds\Logs\APM\RealTime\SolarWinds.APM.ProcessExplorer.log

Anyway, this sounds like issue worth creating support case. Two SNMP walks taken with short delay (few seconds) from affected machine would be useful to better understand what's going on.

If you are on latest version of SAM then switching to polling via linux agent could be workaround. There is good chance that RTPE will work using different polling method. Moreover via Linux agent you should be able to get more info than via SNMP.

akhasheni

Can you find some additional info about this issue in following log when the issue is reproduced?
ProgramData\Solarwinds\Logs\APM\RealTime\SolarWinds.APM.ProcessExplorer.log

2016-12-21 12:52:23,326 [STP SmartThreadPool Thread #39] INFO  SolarWinds.APM.Probes.OneTimeJob.RealTimeProcessesOneTimeJob - Polling processes using SolarWinds.APM.Probes.RealTimeProcessMonitor.SnmpProcessInfoFetcher, winSnmpMode: False2016-12-21 12:52:25,178 [STP SmartThreadPool Thread #39] ERROR SolarWinds.APM.Probes.OneTimeJob.RealTimeProcessesOneTimeJob - Polling processes errorSystem.OverflowException: Value was either too large or too small for a UInt64.   at System.Decimal.ToUInt64(Decimal d)   at System.Convert.ToUInt64(Decimal value)   at SolarWinds.APM.Probes.RealTimeProcessMonitor.SnmpProcessInfoFetcher.GetProcessInfos()   at SolarWinds.APM.Probes.OneTimeJob.RealTimeProcessesOneTimeJob.PollProcesses(IProgressJob progressJob, IProcessInfoFetcher fetcher, Boolean winSnmpMode, TimeSpan pollingInterval, RealTimeProcessesOneTimeJobResultData resultData)   at SolarWinds.APM.Probes.OneTimeJob.RealTimeProcessesOneTimeJob.Probe(OneTimeJobInformation info, RealTimeProcessesOneTimeJobResultData resultData)

Anyway, this sounds like issue worth creating support case. Two SNMP walks taken with short delay (few seconds) from affected machine would be useful to better understand what's going on.

Thank you.

If you are on latest version of SAM then switching to polling via linux agent could be workaround.

Not yet.

lag

Did you find a solution to this problem? I'm bumping into this as well with some RHEL 7.2 servers while trying to use the SNMP process explorer.

aLTeReGo

Have you considered using the Linux agent?

akhasheni

lag wrote:
Did you find a solution to this problem? I'm bumping into this as well with some RHEL 7.2 servers while trying to use the SNMP process explorer.

Not yet.

Linux agent: considering it - but also have to consider security and other potential impacts.

cjfranca

Look this Process Monitor - SNMP error: Value was either too large or too small for a UInt32 - SolarWinds Worldwide, LLC. Help and… and

akhasheni

cjfrancaLook this Process Monitor - SNMP error: Value was either too large or too small for a UInt32 - SolarWinds Worldwide, LLC. Help and… and

The article is about process monitor, not process explorer that's at issue here, and about UINT32, not UINT64. I imagine they're closely related yet there are other strange things about that KB article:

The article has no date - unless I missed it - so it's impossible to say if the latest versions of SAM and SNMP are covered.
The node reports correct memory size in SAM/NPM - via SNMP. How does SAM get it and why can't the same mechanism be used for processes explorers and monitors?
The article says it applies to "All SAM versions" and "All Windows versions".
- The issue is with Linux - so does the article apply here?
- Given it has no date - does it apply to all future SAM and Windows versions? (Nice crystal ball, then - can I borrow it for a little while?)

aLTeReGo

Perform a MIB walk against the device using the SnmpWalk.exe utility located under 'C:\Program Files (x86)\SolarWinds\SnmpWalk.exe'. Open the MIB walk text file and search/find OID '1.3.6.1.2.1.25.5.1.1.2'. Each row starting with '1.3.6.1.2.1.25.5.1.1.2' represents a process running on the machine. The integer column shows the amount of memory consumed by that process. If any of these values are negative (start with a minus symbol) then that would explain the error you are seeing.

leonardo.lopes

Good Morning Alterego

I´m facing this issue, all the values are "0".

What could that means? Is it a problem with the snmp on the server or on NPM?