RAL NGS down due to storage problems - resolved
Dear users,
The storage problems were resovled over the weekend and RAL NGS is now reopened.
Thanks,
Derek Ross
The storage problems were resovled over the weekend and RAL NGS is now reopened.
Thanks,
Derek Ross
RAL NGS down due to storage problems
Due to problems with storage providing NGS home directories at RAL. I have disabled logins to ngs.rl.ac.uk and ngsui03.ngs.rl.ac.uk, and disabled new job starts.
The problem appears to be a memory issue on the individual storage blades causing them to lock up. The vendor is working to provide a permanent fix to the memory issue but for the moment we have reset all blades currently
showing the problem.
Unfortunately, during the reset one of the blades failed an consistency check of its storage and went into a longer consistency check overnight.
That check has now completed, however due to these individual failures the
system as a whole must reconstruct the storage volumes exported to the NGS System. This reconstruction is currently in progress.
Derek Ross
The problem appears to be a memory issue on the individual storage blades causing them to lock up. The vendor is working to provide a permanent fix to the memory issue but for the moment we have reset all blades currently
showing the problem.
Unfortunately, during the reset one of the blades failed an consistency check of its storage and went into a longer consistency check overnight.
That check has now completed, however due to these individual failures the
system as a whole must reconstruct the storage volumes exported to the NGS System. This reconstruction is currently in progress.
Derek Ross
