Mike Christie wrote:
> What firmware do you have, and did yo just start seeing these with a
> target firmware upgrade or a iscsi/ubuntu upgrade?

I can't find it, but I believe we are running version 4.0.1 of the
Equallogic firmware, but we were originally running 3.3.1 and saw
problems with that version as well.

We originally saw these problems on Debian Etch, and they persisted
across our upgrade to Ubuntu Hardy.

> Have you contacted the Equallogic guys? If not I will ping them if
> they do not reply. They might be busy or have not seen the thread here.

We do have a support contract with them and were working on the issue
before. We had thought the problem went away when we moved from Etch to
Hardy. We're planning to get in touch with them again soon, but I'm not
the point of contact on that, so I've asked one of my project's
co-maintainers to do that.

> If you can play around on the box, it would be helpful to run the
> open-iscsi tarball release. Build it with
>
> make DEBUG_TCP=1 make DEBUG_SCSI=1
>
> then do
>
> make DEBUG_TCP=1 make DEBUG_SCSI=1 install
>
> This will overwrite the existing kernel modules and tools with the
> open-iscsi ones, so do not this if you cannot mess around with the box
> and reinstall the ubuntu kernel and tools.
>
> This will dump out a lot of info in /var/log/messages, but we are just
> looking for the part before you see these:
>
> Dec  8 00:50:36 aperture-science kernel: [1010621.595904]
> connection1:0: iscsi: detected conn error (1011)
>
> You might see something about a TCP_CLOSE.

Ok - I've moved all of the virtual machines off of one of our servers,
and it's currently running the latest tarball release of both the
kernelspace and userspace pieces with both debug options enabled.

Our biggest problem when we were trying to diagnose this with Equallogic
was that we didn't have a reliable way of generating the errors. They
don't seem to happen when there's no iSCSI traffic, but they also didn't
occur when we ran a program that generates rapid, random reads and writes.

I'd rather not put production VMs on this server while we're testing. Do
you have any theories on how we could trigger these connection errors,
if they're still occurring?

- Evan

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"open-iscsi" group.
To post to this group, send email to open-iscsi@googlegroups.com
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/open-iscsi
-~----------~----~----~----~------~----~------~--~---

Reply via email to