Bug 9915
Summary: | tar and dd report I/O Error on SCSI Tape while mt seems to work | ||
---|---|---|---|
Product: | [Retired] Red Hat Linux | Reporter: | weinmann |
Component: | mt-st | Assignee: | Trond Eivind Glomsrxd <teg> |
Status: | CLOSED WORKSFORME | QA Contact: | |
Severity: | medium | Docs Contact: | |
Priority: | medium | ||
Version: | 6.1 | CC: | weinmann |
Target Milestone: | --- | ||
Target Release: | --- | ||
Hardware: | i386 | ||
OS: | Linux | ||
Whiteboard: | |||
Fixed In Version: | Doc Type: | Bug Fix | |
Doc Text: | Story Points: | --- | |
Clone Of: | Environment: | ||
Last Closed: | 2000-03-16 15:02:24 UTC | Type: | --- |
Regression: | --- | Mount Type: | --- |
Documentation: | --- | CRM: | |
Verified Versions: | Category: | --- | |
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
Cloudforms Team: | --- | Target Upstream Version: | |
Embargoed: |
Description
weinmann
2000-03-02 09:22:28 UTC
We exchanged the tape with another one from a different vendor but the behaviour is the same. mt status, mt rewind and mt retension work fine. But mt setblk xxx, tar and dd report /dev/st0 I/O-Error. All of status/retension/rewind do not read data from the media, while tar/dd do, so the behavior is consistent. Are there error messages in /var/log/messages? Is there *any* tape that can be read by the drive? Is there any tape that can be *written* by the drive? On dd if=xxx of=/dev/st0 /var/log/messages has an entry Mar 10 14:50:17 altair kernel: st0: Write not multiple of tape block size. Mar 10 14:51:17 altair last message repeated 2 times I checked with different blocksizes: (bs=0, bs=1, bs=64, bs=512, bs=1024) but it is all the same. On tar cfv /dev/st0 xxx there is no entry in /var/log/messages. But if I issue afterwards mt rewind and then tar tfv /dev/st0 the tar command responds with tar: Hmm, das sieht nicht wie ein ;tar+-Archiv aus. tar: Springe zum Kopfteil der ndchsten Datei. . Translation: Doesn't look like a tar archive. Jumping to next File. We tried to write with different tapes (but all same vendor/type). But the result was the same. We have the same type of tape also in another Netfinity 5500 (no M20) running DLD Linux and it works fine since nearly 1 1/2 years. Regarding reading tapes: Though we have a tape device that works in another Netfinity 5500 I cannot change the backup procedures there and I have no access to that tape device or the tapes. This server is running in production and I am not allowed to use it for some test purposes. Stuff like cabling and termination an external HW support company and myself already verified. We installed an Adaptec SCSI controller in addition and attached the tape device to it. Now we can work with the tape device. Though this is a workaround and not a fix it solves our issue to work with the tape device. So we will not continue to firther investigate and the request is closed from our side. I believe the culprit is the ServeRAID driver not handling tape drives. We added a Adaptec SCSI adapter, as well, based on the last posting here. That works fine. The only difference was the adapter/driver. Hope this helps. |