thor918
Enthusiast
Enthusiast

cpu spikes on a poweredge 850

hi there.

I just finished up putting togheter a poweredge 850 with a perc5i sas raid card with two sata disks in raid 1.

The system seems fine, all is green in health. it has only 1GB ram at the moment. But runs just fine with that.

I just noticed that when I look at the preformance window for my cpu, there are regular cpuspikes.

it seems they accour every 10min

http://home.no.net/thor918/vmware/spikes.jpg

anyone have any clue why I get these spikes with a average 40%?

the spikes accoure even if no virtual machines are running.

0 Kudos
52 Replies
nick_couchman
Immortal
Immortal

Download the RCLI tools and use the resxtop utility to determine what process is spiking the CPU.

0 Kudos
thor918
Enthusiast
Enthusiast

I can't find "resxtop" in the windowspackage, I can find it in the linux version however I'm stopped there with a gcc error.

do you know where it is in the windows version?

0 Kudos
Dabj
Contributor
Contributor

I have the same problem on a PowerEdge 2950, Around every 10 minutes. Have seen more forum entries but no solution. Just before the spike a extra process launches: sfcdb.3968279. Has anyone a idea?

Thnx,

0 Kudos
thor918
Enthusiast
Enthusiast

hmm I see.

Dabj, did you by the way run the top command in a cli or on the box itself, when you found the process?

here is something similar:

http://communities.vmware.com/thread/165406?tstart=80

If there are any other topics on the same subject, please post a link.

oh.

http://communities.vmware.com/thread/136040?tstart=15

it seems I'm correct with top command script is left outside the windows package!

0 Kudos
thor918
Enthusiast
Enthusiast

Download the RCLI tools and use the resxtop utility to determine what process is spiking the CPU.

okey. I finaly got the resxtop working in a linux shell. it isn't hard to spot the spike in the top utility.

I have attached a top-screenshot of the cpu spike

0 Kudos
nick_couchman
Immortal
Immortal

Looks like there's another thread on the issue:

http://communities.vmware.com/message/957023

Seems to indicate maybe something related to hardware health monitoring? I don't seem to see the same behavior on my whitebox machiines...

0 Kudos
thor918
Enthusiast
Enthusiast

Looks like there's another thread on the issue:

http://communities.vmware.com/message/957023

Seems to indicate maybe something related to hardware health monitoring? I don't seem to see the same behavior on my whitebox machiines...

not sure it's exactly the same issue, because my server does work non stop.(from what I can see)

anyways.. what is purpouse of sfcb prosess?

I just did an adjustment of the resources for the sfcb prosess:

http://home.no.net/thor918/vmware/vmware-sfcb-adjustment.jpg

configuration->"system resource allocation"->advanced->sfcb tree

limit set to 475Mhz for the sfcb prosess.

this is how the spikes looks now after the change:

http://home.no.net/thor918/vmware/vmware-sfcb-adjustment_dia.jpg

the spikes is reduced to about 10% avrage, but seems to take about the same time.

0 Kudos
thor918
Enthusiast
Enthusiast

Download the RCLI tools and use the resxtop utility to determine what process is spiking the CPU.

okey. I finaly got the resxtop working in a linux shell. it isn't hard to spot the spike in the top utility.

I have attached a top-screenshot of the cpu spike

hmm

just wondering.

look at my screenshoot of resxtop, and look at idle. it sums up to about 200%.

is it because it's a dualcore D prosessor?

I'm looking on my other esxi-installation on another server, poweredge 1850:

http://www.padova.infm.it/Calcolo/Download/1850_specs.pdf

That one also looks like it's dual core prosessor. this one however does not have the spikings problem, and it does list max 100% in resxtop.

0 Kudos
thor918
Enthusiast
Enthusiast

/var/log/messages

Sep 2 10:48:53 Hostd: Task Completed : haTask--vim.SimpleCommand.Execute-1014

Sep 2 10:48:57 LSIESG: LSIESG:INTERNAL :: StorelibManager::fireStorelibCommand - caller StorelibManager::getConnectorInfo, ProcessLibCommandCall failed, rval = 0x2

Sep 2 10:48:57 sfcbd: INTERNAL StorelibManager::fireStorelibCommand - caller StorelibManager::getConnectorInfo, ProcessLibCommandCall failed, rval = 0x2

Sep 2 10:48:57 LSIESG: LSIESG:INTERNAL :: StorelibManager::discover - DatadiscoveryfailedforConnector;Errorcode=2

Sep 2 10:48:57 sfcbd: INTERNAL StorelibManager::discover - DatadiscoveryfailedforConnector;Errorcode=2

hmmm

0 Kudos
thor918
Enthusiast
Enthusiast

another thread about the same thing!

http://communities.vmware.com/thread/165406?tstart=0

0 Kudos
thor918
Enthusiast
Enthusiast

I was hoping that others with the same problem could try to add data to what's going on.

It might get a better chance to be fixed if the vmware team has something to work with.

still even if this is free, and paid for support, I would think that vmware does read up on this forum.

also we could post a bug report, but then we need much more data on the problem.

0 Kudos
Dabj
Contributor
Contributor

Sorry was some what occupied.

entries in var/log/messages every 10 minutes:

Sep 8 10:04:51 LSIESG: LSIESG:INTERNAL :: StorelibManager::fireStorelibCommand - caller StorelibManager::getConnectorInfo, ProcessLibCommandCall failed, rval = 0x2

Sep 8 10:04:51 sfcbd: INTERNAL StorelibManager::fireStorelibCommand - caller StorelibManager::getConnectorInfo, ProcessLibCommandCall failed, rval = 0x2

Sep 8 10:04:51 LSIESG: LSIESG:INTERNAL :: StorelibManager::discover - DatadiscoveryfailedforConnector;Errorcode=2

Sep 8 10:04:51 sfcbd: INTERNAL StorelibManager::discover - DatadiscoveryfailedforConnector;Errorcode=2

Sep 8 10:04:51 LSIESG: LSIESG:INTERNAL :: PersistentJob::createRepositoryObject: found 1 MegaRAIDHBAs

Sep 8 10:04:51 sfcbd: INTERNAL PersistentJob::createRepositoryObject: found 1 MegaRAIDHBAs

Sep 8 10:04:51 LSIESG: LSIESG:INTERNAL :: Assert failed, CheckConsistencyJobProvider::populateKeyAttributes: CheckConsistencyDetails Object is NULL (../SBMA_CIM_SMI12_Internal/cim/prov/smi12/hhr/jobcontrol/CheckConsistencyJobProvider.cc:105)

Sep 8 10:04:51 sfcbd: INTERNAL Assert failed, CheckConsistencyJobProvider::populateKeyAttributes: CheckConsistencyDetails Object is NULL (../SBMA_CIM_SMI12_Internal/cim/prov/smi12/hhr/jobcontrol/CheckConsistencyJobProvider.cc:105)

Sep 8 10:04:51 LSIESG: LSIESG:INTERNAL :: Caught SBMA Exception.....

Sep 8 10:04:51 sfcbd: INTERNAL Caught SBMA Exception.....

Sep 8 10:04:51 LSIESG: LSIESG:INTERNAL :: EventDistributor::propagateListenerNotifications: caught SbmaException: Exception: 1 (0x1): Caught SBMA Exception : IndicationProvider::processCreationEvent:(../SBMA_CIM_Framework/cim/prov/IndicationProvider.cc:184)

Sep 8 10:04:51 sfcbd: INTERNAL EventDistributor::propagateListenerNotifications: caught SbmaException: Exception: 1 (0x1): Caught SBMA Exception : IndicationProvider::processCreationEvent:(../SBMA_CIM_Framework/cim/prov/IndicationProvider.cc:184)

Sep 8 10:04:51 LSIESG: LSIESG:INTERNAL :: EventDistributor::processTransactionToListeners: caught SbmaException: Exception: 1 (0x1): Caught SBMA Exception : IndicationProvider::processCreationEvent:(../SBMA_CIM_Framework/cim/prov/IndicationProvider.cc:184)

Sep 8 10:04:51 sfcbd: INTERNAL EventDistributor::processTransactionToListeners: caught SbmaException: Exception: 1 (0x1): Caught SBMA Exception : IndicationProvider::processCreationEvent:(../SBMA_CIM_Framework/cim/prov/IndicationProvider.cc:184)

Sep 8 10:04:51 LSIESG: LSIESG:INTERNAL :: EventDistributor::processReadyTransactions: caught exception in listener callback, transaction 29: Exception: 1 (0x1): Caught SBMA Exception : IndicationProvider::processCreationEvent:(../SBMA_CIM_Framework/cim/prov/IndicationProvider.cc:184)

Sep 8 10:04:51 sfcbd: INTERNAL EventDistributor::processReadyTransactions: caught exception in listener callback, transaction 29: Exception: 1 (0x1): Caught SBMA Exception : IndicationProvider::processCreationEvent:(../SBMA_CIM_Framework/cim/prov/IndicationProvider.cc:184)

rg (END)

Thanks for the patience.

0 Kudos
thor918
Enthusiast
Enthusiast

thanks! looks like your errors are like mine.

what diskcontroller?

how many disks?

and are there disk on all ports?

if scsi, do you have termination dongle on diskports not used`?

0 Kudos
Dabj
Contributor
Contributor

Hi thor,

Dell 2950 with PERC5/i FW 5.2.1-0067

4 disks in raid 10

Storage adapters tab states 631xESB / 632xESB IDE controller | vmhba0 Block scsi for CDROM. But there is also a USB StorageController entry vmhba32 SCSI vmhba32:0:0 which also states cdrom, I rackin this is the virtual cdrom from DELL.

'out of the box' was hosting win2k3 for the last 2 years. So nothing has changed like termination etc.

And now for the frustrating part: this morning for the last response i enabled SSH which did a stop / start of the sfcbd processes after which i gave a sighup of inetd ............................. NO SPIKES !@!@#$%

At this moment i can't reboot the server some blokes have there virtual 'test' servers running. I saw a post earlier that after a reboot the spikes returned.

Grtz,

0 Kudos
thor918
Enthusiast
Enthusiast

hmm I see we have similar hardware.

I have exactly the same controllercard with exactly the same firmware.

I read the same thing about rebooting, however in my case the spikes are there no matter how many times I restart.

I have not tried restarting the controller service like you did, I have not enabled the ssh feature. did the spikes return after a while?

0 Kudos
Dabj
Contributor
Contributor

A few days later (sorry no reboot) and NO SPIKES.

Grtz,

0 Kudos
Dabj
Contributor
Contributor

Found a time slot to reboot !@@##$

Spikes are back. Going to restart sfcb processes. Let you know wat happens.

Grtz,

0 Kudos
thor918
Enthusiast
Enthusiast

hmm,

can you just post exactly what commands you issued on your ssh connection?

0 Kudos
Dabj
Contributor
Contributor

Last time i enterd services.sh restart after that i gave kill -HUP process number of inetd.

No when i do the same the spike instantly returns. Tried it several times but no luck.

Grtz,

0 Kudos