xfs
[Top] [All Lists]

Courier Imap 3.0.7 rpmbuild fails with xfs->lvm2->md1

To: linux-xfs@xxxxxxxxxxx
Subject: Courier Imap 3.0.7 rpmbuild fails with xfs->lvm2->md1
From: Kai Leibrandt <k_leibrandt@xxxxxxxxxxx>
Date: Wed, 08 Sep 2004 09:21:12 +0200
Sender: linux-xfs-bounce@xxxxxxxxxxx
User-agent: Mozilla Thunderbird 0.7.3 (Windows/20040803)
Hi all,

I am trying to build courier-imap-3.0.7 using rpmbuild -ta as per instructions at http://www.inter7.com/courierimap/courierimap.html but unfortunately this fails with the following result:

**********
/usr/bin/make  check-am
make[2]: Entering directory `/var/export/home/kailee.net/kai/rpm/BUILD/courier-imap-3.0.7/maildir'
/bin/sh ./testsuite 2>&1 | cmp - ./testsuite.txt
LC_ALL=C; export LC_ALL; ./maildirkwtest | cmp -s - ./maildirkwtest.txt
LC_ALL=C; export LC_ALL; ./maildiraclttest
testmaildiraclt.c(145): sanity check failed.
make[2]: *** [check-am] Error 1
make[2]: Leaving directory `/var/export/home/kailee.net/kai/rpm/BUILD/courier-imap-3.0.7/maildir'
make[1]: *** [check] Error 2
make[1]: Leaving directory `/var/export/home/kailee.net/kai/rpm/BUILD/courier-imap-3.0.7/maildir'
make: *** [check-recursive] Error 1
error: Bad exit status from /var/tmp/rpm-tmp.27389 (%build)


RPM build errors:
    Bad exit status from /var/tmp/rpm-tmp.27389 (%build)
**********

I've tried opening up the spec and building from that but it gives me the exact same result.

System is FC2, kernel 2.6.8-1.521, rpm 4.3.1, gcc 3.3.3, all filesystems are xfs->lvm2->md1 raid.

The same system will happily rpmbuild all the source _if the fs in question is not on lvm_, ie. xfs->md1.

This has been posted to the Courier-Imap mailing list too, and I got the following info from there:
**********
> Sorry for following up to my own post; I have tried doing the same
> thing on an identical setup without lvm (FC2 on xfs->software raid1
> instead of xfs->lvm2->software raid1) and there it works. Can this
> have anything to do with the error I get?


Yes, although I don't know what the actual problem is.

The portion of the code in question makes some internal library calls whose intended effect is to cause certain content to be written out to certain files. It then verifies whether the expected files actually exist.

However, I don't see how LVM could possibly be a factor. LVM should not make any difference, as long as you're using the same filesystem on top. But, say, if it's ext3 in one case, and something else (reiser? xfs?) in the other case, then you'll might be onto something.
**********

Sounds a little like the grub-bug - data not being written?

Any help will be greatly appreciated.


Kai.


<Prev in Thread] Current Thread [Next in Thread>