This blog deals with a problem which was ongoing from October 2010 until February 2011. It looks like it has finally settled down a bit (in other words a workable fix has been found), but it’s still not too widely publicized.
A posting on Cisco’s NetPro states that there are firmware issues with the hard drives in 7825-I4, 7828-I4m and 7816-I4 servers. The posting can be found here:
The upshot of the post is that a technician knows that a server is in the middle of having the problem caused by this firmware if the following error is seen on the monitor:
EXT3-fs error (device sda6) in start_transaction: Jornal has aborted
Also, if the technician can log into an SSH session, the following information will be displayed:
Java.io.FileNotFoundException: /var/log/active/platform/log/cli.bin (Read-only
file system) at java.io.RandomAccessFile.open(Native Method)
log4j:ERROR No output stream or file set for the appender named [CLI_LOG]
The issue is something that has been seen on specific servers with hard drives listed as model
“WD2502ABYS-23B7A”, and a firmware version of “3B04”. The model and firmware versions can be seen by logging into the CLI of the server and issuing the “show hardware” command.
Again, looking at the NetPro document is worthwhile because it goes into a lot of detail about the problem, and because the document is being updated frequently as IBM and Cisco work together to arrive at a resolution that they feel is completely satisfactory.
We encountered this issue both just before the NetPro article was posted, as well as just after. If you are interested in reading our (rather long) narrative about our experiences click here for the full version.