Jody - <br><br>Not a problem to add a section in the Lustre manual re: supported upgrades. However, as we do not currently sync Lustre releases with releases of the Lustre manual, this wouldn't help us to address the issue that Jeff poses.
<br><br>For future Lustre releases, could we add a supported upgrade/interop statement to the change log? I'll also add this content to the manual (which will have more value once Lustre s/w and manual releases are synchronized).
<br><br>Sheila<br><br><div><span class="gmail_quote">On 10/12/07, <b class="gmail_sendername">Jody McIntyre</b> <<a href="mailto:scjody@clusterfs.com">scjody@clusterfs.com</a>> wrote:</span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Hi Jeff,<br><br>On Fri, Oct 12, 2007 at 12:01:04PM -0400, Jeff Blasius wrote:<br><br>> I've seriously trekked through the lustre documentation and haven't<br>> found an answer regarding this. Is there an official policy regarding
<br>> interoperability among different versions of various lustre<br>> components?<br><br>By coincidence, I just sent information about this to our documentation<br>team. It should eventually reach the manual. Here it is:
<br><br>---<br>Our supported upgrades are from one minor version to another, for<br>example 1.4.10 to 1.4.11 or 1.6.2 to 1.6.3, and also from the latest<br>1.4.x version to the latest 1.6.x version, so 1.4.11 to 1.6.3 is<br>
supported.<br><br>We also support downgrades within the same ranges. For example, if you<br>upgrade from 1.6.2 to 1.6.3, you can also downgrade to 1.6.2 (but a<br>fresh install of 1.6.3 is _not_ guaranteed to be downgradeable.)
<br><br>Note that other combinations will work and we support them in specific<br>cases when requested by customers, but the ranges above will always<br>work.<br>---<br><br>> For example (and I'm sure other groups are in the same boat here),
<br>> it's relatively painless to perform a rolling upgrade to the lustre<br>> clients, but upgrading the OSS or MDS takes more convincing. Is it OK<br>> for me to run a patched but 1.6.0 based OSS with a 1.6.3
client? In<br>> this case all of the lustre components (kernel, lustre, ldiskfs) are<br>> the same version for each host. Similarly, is it OK to run a lustre<br>> kernel version out of sync with the userland tools? For example a
<br>> 1.6.0 kernel with a 1.6.3 lustre build on the same host?<br><br>Not necessarily. You should do a rolling upgrade to 1.6.1, then 1.6.2,<br>then 1.6.3. Upgrading will be easier if you stay more current - 1.6.0<br>
is fairly old at this point.<br><br>Having said that, I believe 1.6.0 and 1.6.3 is actually something that<br>will work, but I'm not 100% certain of this, so I'll allow others to<br>correct me.<br><br>Cheers,<br>Jody
<br><br>> I understand that many of these combinations do in fact work, I'm more<br>> interested if they're likely to lead to data corruption or client<br>> evictions. I'm not sure how often incompatibilities arise, but if it's
<br>> relatively rare, it would be useful if that was announced on the<br>> change log. Of course if there's a serious "Do at your own risk"<br>> policy that would also be useful to know.<br>><br>
> Thank You,<br>> jeff<br>><br>> --<br>> Jeff Blasius / <a href="mailto:jeff.blasius@yale.edu">jeff.blasius@yale.edu</a><br>> Phone: (203)432-9940 51 Prospect Rm. 011<br>> High Performance Computing (HPC)
<br>> Linux Systems Design & Support (LSDS)<br>> Yale University Information Technology Services (ITS)<br>><br>> _______________________________________________<br>> Lustre-discuss mailing list<br>>
<a href="mailto:Lustre-discuss@clusterfs.com">Lustre-discuss@clusterfs.com</a><br>> <a href="https://mail.clusterfs.com/mailman/listinfo/lustre-discuss">https://mail.clusterfs.com/mailman/listinfo/lustre-discuss</a><br>
><br><br>--<br><br>_______________________________________________<br>Lustre-discuss mailing list<br><a href="mailto:Lustre-discuss@clusterfs.com">Lustre-discuss@clusterfs.com</a><br><a href="https://mail.clusterfs.com/mailman/listinfo/lustre-discuss">
https://mail.clusterfs.com/mailman/listinfo/lustre-discuss</a><br><br></blockquote></div><br>