Location: Ask the Storage Experts

Discussion: HELP, dell support case SR# 810122257 open all month and still open!Reported This is a featured thread

Showing 6 posts
resadeealvord
resadeealvord
HELP, dell support case SR# 810122257 open all month and still open!
Feb 1 2010, 3:32 PM EST | Post edited: Feb 1 2010, 3:32 PM EST
Hi,
We're experiencing serious performance issues with a Dell R900 running Windows 2008 attached to a MD3000. We are trying to copy data from another server to this storage and the system is generating a series of eventlogs which brings the system to it's knees. We have had an open case with Dell tech support all month (SR# 810122257) . Can any of the Dell storage experts please take a look at this case and assist us?
thanks!
Resa Alvord
Cornell Center for Advanced Computing
Ithaca, NY
Do you find this valuable?    
Keyword tags: best practice experts storage
KongY@Dell
KongY@Dell
1. RE: HELP, dell support case SR# 810122257 open all month and still open!
Feb 2 2010, 9:53 AM EST | Post edited: Feb 2 2010, 9:53 AM EST
Resa,

Please share details of your performance issue. What are the events that event viewer / MDSM sees? Also, what's your MD3000 configuration i.e. how many disk groups, what RAID levels, how many spindles in each disk group? What are your R/W latencies to your RAID volumes, avg. transfer size per R/W, CPU utilization, and # of interrupts per second?
Do you find this valuable?    
resadeealvord
resadeealvord
2. RE: HELP, dell support case SR# 810122257 open all month and still open!
Feb 2 2010, 2:04 PM EST | Post edited: Feb 2 2010, 2:04 PM EST
Hi,
THanks for responding. We're on the phone with someone from the DELL escalation team now. I'll quickly say that the MDSM shows optimal status and is not throwing any errors. All errors are thrown by Windows : raid controller reset: Reset to device, \Device\RaidPort2 (event 129 lsi_sas) . I'll post more later.
Resa
Do you find this valuable?    
resadeealvord
resadeealvord
3. RE: HELP, dell support case SR# 810122257 open all month and still open!
Feb 5 2010, 2:44 PM EST | Post edited: Feb 5 2010, 2:44 PM EST
CASE RESOLVED: It turns out the raid controller was 'stuck' in a battery learn cycle. Power cycling did not 'fix' this. The fix was to manually disable and re-enable the read and write caching for the controller... we are now successfully copying files to the volume without any performance degradation! SIGH! We went through a lot of hoopla to finally find this fix. The MDSM showed everything optimal throughout, the only clues were the Windows errors. I hope this makes it to a Dell Support Document for the MD3000 so others can fix more quickly. Do you find this valuable?    
resadeealvord
resadeealvord
4. RE: HELP, dell support case SR# 810122257 open all month and still open!
Feb 5 2010, 3:10 PM EST | Post edited: Feb 5 2010, 3:10 PM EST
FWIW, here are the commands used to fix the problem with the MD3000 controller:

smcli -n web-storage -c "set allVirtualDisks readCacheEnabled=FALSE writeCacheEnabled=FALSE;"
smcli -n web-storage -c "set allVirtualDisks readCacheEnabled=True writeCacheEnabled=True;"
SMcli -n web-storage -c "set storageArray learnCycleDate daysToNextLearnCycle=0;"

Resa
Do you find this valuable?    
KongY@Dell
KongY@Dell
5. RE: HELP, dell support case SR# 810122257 open all month and still open!
Feb 5 2010, 5:13 PM EST | Post edited: Feb 5 2010, 5:13 PM EST
Great news Resa!

Thanks for sharing the fix and your experience with our TechCenter community.
Do you find this valuable?    

Related Content

  (what's this?Related ContentThanks to keyword tags, links to related pages and threads are added to the bottom of your pages. Up to 15 links are shown, determined by matching tags and by how recently the content was updated; keeping the most current at the top. Share your feedback on Wetpaint Central.)