Multiple messages "FATAL: Failed to read file! file: 128 pos: 2588672 errno = 0 ()" when testing industrial Compact Flash

Bug #1332579 reported by virgil kuruthers
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
sysbench
New
Undecided
Unassigned

Bug Description

Not sure if this is a bug, or a legit. problem found by sysbench.

I am testing a new industrial rated compact flash memory on an embedded Atom system. We are using Debian Squeeze. I ran a test of tests with this command:

while true; do sysbench --num-threads=16 --test=fileio --file-total-size=3500M --file-test-mode=rndrw prepare; sysbench --num-threads=16 --test=fileio --file-total-size=500M --file-test-mode=rndrw run; sync; sysbench --num-threads=16 --test=fileio --file-total-size=3500M --file-test-mode=rndrw cleanup; done

and let it run for about 3 weeks.

When I returned I saw a lot of messages like these on the console output

FATAL: Failed to read file! file: 127 pos: 2965504 errno = 0 ()
FATAL: Failed to read file! file: 127 pos: 540672 errno = 0 ()
FATAL: Failed to read file! file: 130 pos: 655360 errno = 0 ()
FATAL: Failed to read file! file: 129 pos: 2441216 errno = 0 ()
FATAL: Failed to read file! file: 129 pos: 475136 errno = 0 ()
FATAL: Failed to read file! file: 128 pos: 2588672 errno = 0 ()
FATAL: Failed to read file! file: 126 pos: 2588672 errno = 0 ()
FATAL: Failed to read file! file: 127 pos: 2899968 errno = 0 ()
FATAL: Failed to read file! file: 125 pos: 3309568 errno = 0 (T¦v¦)
FATAL: Failed to read file! file: 126 pos: 49152 errno = 0 (¦v¦0R0¦¦¦v¦R0¦¦u¦¦Q0¦T¦Q0¦T¦v¦)

Does this indicate an actual problem found with the Compact Flash hardware, or could it be a bug in sysbench? This answer is very important as we will need to deploy 1000s of these parts and they need to be reliable.

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.