PR_APTPL_BUF_LEN buffer size

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hello,

we have an issue with a higher number of Hyper-V hosts using one APTPL-enabled LIO iSCSI target LUN.
In particular, it's not possible to add more than 6 Hyper-V hosts to CSV cluster and target dmesg
shows the following errors:

[60451129.483580] Unable to update renaming APTPL metadata
[60451129.483585] Unknown transport error for CDB 0x5f: 0
[60451129.483948] Unable to update renaming APTPL metadata
[60451129.483952] Unknown transport error for CDB 0x5f: 0

Looking into reservations code, I guess that we're hitting PR_APTPL_BUF_LEN buffer size. With 6
hosts and 4 iSCSI sessions from every initiator, the /var/target/pr/aptpl file size is very close to
the hardcoded 8192 bytes and there's no room for another host. When we reduce the number of active
sessions of some hosts to 2, it's possible to add the 7th host.

Note that we automatically enforce aptpl=1 for all PRs in target_scsi3_emulate_pr_out() by a custom
patch. Without enforced APTPL, we observed random CSV stability issues during LIO HA target
failovers and other maintenance processes. Would the problem be related to it?

Otherwise, can we workaround it just by increasing the PR_APTPL_BUF_LEN to e.g. 80 kB or is it a
more complex problem?

We use kernels 3.7.10+ and 3.15.10+, but the 8 kB buffer limit seems to be in latest code too.

Thank you,

Martin Svec

--
To unsubscribe from this list: send the line "unsubscribe target-devel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html




[Index of Archives]     [Linux SCSI]     [Kernel Newbies]     [Linux SCSI Target Infrastructure]     [Share Photos]     [IDE]     [Security]     [Git]     [Netfilter]     [Bugtraq]     [Yosemite News]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Linux RAID]     [Linux ATA RAID]     [Linux IIO]     [Device Mapper]

  Powered by Linux