Ralf G. R. Bergs wrote: On Mon, 2002-01-07 at 12:51, Steve Lord wrote: [...] Sean, can you see if the attached patch against the current cvs tree fixes the problem for you? So far I have not been abl
[...] whether new I see. Sorry, my fault. Do you have any idea what else I could do to further investigate *my* problem? Thanks, Ralf -- Verkaufe Original-BMW-Raeder: L I N U X .~. http://adsl-bergs.
Ralf G. R. Bergs wrote: I see. Sorry, my fault. Do you have any idea what else I could do to further investigate *my* problem? Thanks, Ralf You are in the 'queue', unfortunately, there are not many h
OK, I think I will put the code in and wait to hear back from you on this one. Steve -- Steve Lord voice: +1-651-683-3511 Principal Engineer, Filesystem Software email: lord@xxxxxxx
I'm getting some random file corruption on one of my XFS LVM volumes, leading to things like an emacs binary that is filled with runs of NULs and pieces of deleted files. I've been able to build succ
How much memory do you have on this machine, and how many cpus? And can you tell if the build is doing things in parallel or is just single threaded? I don't think there is much point going back to e
begin Steve Lord quotation: 256M: two 128M DIMMs. I ran two passes of memtest a couple days ago with no errors. I have one CPU, a PII-450. The kernel is a UP build, with Robert Love's preempt patch a
begin Sean Neakums quotation: I'm now running a fresh build of 2.4.17, without the preempt patch. emacs reliably fails to build on the XFS volume, and builds successfully on the ext2 volume. -- /////
Doing a non lvm build here, lets see if the that works. I have my suspicions that lvm is going to be the magic factor here. Steve -- Steve Lord voice: +1-651-683-3511 Principal Engineer, Filesystem S
begin Steve Lord quotation: I'm using the LVM as shipped in SGI's CVS tree, which claims to be "1.0.1-rc4(ish)". As I recall, that's the version being shippedin stock 2.4.17. Is there a newer version
No, I was refering to XFS on top of lvm being a possible culprit, independent of lvm version. I completed an emacs build - surprisingly quickly, but I want to try a different version, I am building r
begin Steve Lord quotation: I've just realised that my swap is divided between two partitions. I've taken one and created an XFS filesystem on it. Building emacs now. -- ///////////////// | | The spa
begin Sean Neakums quotation: Weird, it failed the with the same errors I was getting using the XFS LVM volume: make[2]: ** No rule to make target `/datacomp/emacs21-21.1/src/../lisp/register.elc', n
FYI, file compression is extremely sensitive to marginal ram. If unpacking has anything to do with it at all, and not the source of the unpack or destination, you should double-check ram with memtest
begin D. Stimits quotation: I ran two passes of memtest-86 a few days ago, after I had that XFS filesystem shutdown. I'm not getting any errors at the unpack stage. I mentioned that I was unpacking f
Since I am building on a redhat based box, I am doing rpm builds not debian package builds, and the rpm version which appears friendly to this box is emacs-20 not emacs-21. Having said that, emacs is
This is probabily unrelated & OT but just straight XFS on LVM seems OK in the fact that I have not seen any problems with it as yet (but then again I have not been looking for problems like file corr
begin Steve Lord quotation: I don't even get as far as generating packages when I build on XFS; the part I mentioned about temacs re-appearing mysteriously happens during the actual emacs21 build. I
Following up to myself, dropping the system memory to 128M seems to have triggered the problem here, so yes there does seem to be an XFS bug here. Steve -- Steve Lord voice: +1-651-683-3511 Principal
Yes, I have seen the thread, I have just not had the time to look into it yet. Your last message said that the xfs snapshot creation was failing, but with an ext3 stack trace..... You also mention ou