-
Notifications
You must be signed in to change notification settings - Fork 68
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
LTFS Mount read only - possible reason #448
Comments
I'm not sure your first log is produced by which step.
But it is cerated by step5, LTFS detected a permanent write error on step4 and make it read only at step4. So the tape was mounted as R/O at step5. The first log says the tape was mounted as R/O like below. The log produced at step3 and step4 is required if you want to investigate the reason of the permanent write error. LTFS shall generate drive dump files because of the permanent write error. You need to contact to IBM for more investigation.
|
Hi @piste-jp-ibm thanks for this quick answer. I will try to find back the previous umount for this tape, and try to provide the log. I suspect that the tape(s) for some unknown reason are full at one stage (before raising its total capacity), and then there is no space anymore to write the index. Or something like this. Do you believe this could be a root cause? I come back to you with the info. |
I don't think so. The TA reported from library might not be related into this problem. Tape device is basically append device, so a permanent write error happens at a position the drive cannot write any following data. (Of course, we can issue another write at the error position and might have good response luckily and continue to write. But LTFS never does such kind of handling, because such kind of write error might cause bigger problem at read side later!) In this case, LTFS might have a permanent write error in the step3 or step4. And LTFS make the tape R/O because of the permanent write error. This is the design of LTFS and LTFS works fine I believe. Additional failure analysis is required by the IBM engineers to find out the root cause of the permanent write error. |
I found back the previous LTFS log, and we have found i believe interesting information:
And specially this extract:
If you have some explanation of what happened , this could help to troubleshoot the problem. |
It looks everything works as expected.
|
If you want to know more detail of the permanent write error, please contact to IBM support with the dump files, LTFS doesn't know the symptom of write error and it is not a scope of LTFS. LTFS just takes actions against error code (SCSI sense key). In this case, action is
|
Describe the bug
I'm facing to a case where the LTFS mount results with read-only mount.
This happened multiple time with different LTO9. I never get this before, and i try to understand what's happened.
To Reproduce
The steps to reproduce are the following:
Trace Log
Here is the log of a tape where i have the problem:
Another one without the problem (yet), i just did a LTFS format:
Desktop (please complete the following information):
LTFS 2.4.4.1 is used
The text was updated successfully, but these errors were encountered: