Simpler Navigation coming for Servers and Operating Systems
Coming soon: a much simpler Servers and Operating Systems section of the Community. We will combine many of the older boards, and you won't have to click through so many levels to get at the information you need. If you are looking for an older board and do not find it, check the consolidated boards, as the posts are still there.
Operating System - Tru64 Unix
cancel
Showing results for 
Search instead for 
Did you mean: 

Emlex HBA message :- Warning, ring[0].hash[0] iocbhdr

A.W.R
Frequent Advisor

Emlex HBA message :- Warning, ring[0].hash[0] iocbhdr

Hi,

I have emulex FC HBAs in my Alpha servers with Tru64UNIX v5.1B. From time to time I get the following messages:-

Aug 4 15:56:48 vmunix: EMX[0]: Warning, ring[0].hash[0] iocbhdr 0xfffffc01419aac88 stalled for 65 seconds!
Aug 4 15:56:48 vmunix: ccb 0xfffffc01419aae78 0/20
Aug 4 15:56:48 vmunix: EMX[0]: Warning, ring[0].hash[1] iocbhdr 0xfffffc015ef33708 stalled for 65 seconds!
Aug 4 15:56:48 vmunix: ccb 0xfffffc015ef338f8 0/30

If anyone can let me know what would be causing this it would be greatly appreciated.

Thanks
Andrew
2 REPLIES
Martin Moore
HPE Pro

Re: Emlex HBA message :- Warning, ring[0].hash[0] iocbhdr

These are informative messages that were introduced in V5.1B-4 to indicate when FC disk I/O is backing up. If it happens at consistent times (such as during filesystem backups) it could just be a transient condition indicating extremely heavy I/O. Alternatively, it could indicate a problem somewhere in the I/O subsystem. You could check the error log to see if anything was logged around the same time.

Martin
I work for HP
A quick resolution to technical issues for your HP Enterprise products is just a click away HP Support Center Knowledge-base
See Self Help Post for more details

DCBrown
Frequent Advisor

Re: Emlex HBA message :- Warning, ring[0].hash[0] iocbhdr

As previously indicated, these are warnings... until it starts complaining about errors.

Note tat there was a hole in the logic that was corrected in later versions of the driver concerning tape io. Normal disk io has a 60 second timeout and a lot of tape io has several minute to 90 minute timeout -- but some tape io has < 250 second timeout.

The stall config limit is set at 60 seconds I think, but if tape io has a 80 second timeout and it takes 75 seconds to complete the io then the driver will issue warnings about stalled io when there really isn't any.

If there are no tapes on the san... worry about slow io otherwise bump the stall config warning time up to 90-120 seconds.