Running kernel 2.6.25-r1 (or -r2) with openafs (at least 1.4.7 or 1.4.7-rc?) freezes openafs. Initially the system mounts /afs OK but accessing it by ls /afs/... or klog or df ... freezes openafs. Only rebooting helps. Reproducible: Always emerge --info Portage 2.1.5_rc6 (default/linux/x86/2008.0, gcc-4.2.3, glibc-2.7-r2, 2.6.25-gentoo-r2 i686) ================================================================= System uname: 2.6.25-gentoo-r2 i686 Intel(R) Pentium(R) III CPU family 1266MHz Timestamp of tree: Tue, 06 May 2008 07:45:01 +0000 ccache version 2.4 [disabled] app-shells/bash: 3.2_p39 dev-java/java-config: 1.3.7, 2.1.6 dev-lang/python: 2.5.2-r2 dev-python/pycrypto: 2.0.1-r6 dev-util/ccache: 2.4-r7 sys-apps/baselayout: 2.0.0 sys-apps/openrc: 0.2.3 sys-apps/sandbox: 1.2.20_alpha2-r1 sys-devel/autoconf: 2.13, 2.61-r1 sys-devel/automake: 1.4_p6, 1.5, 1.6.3, 1.7.9-r1, 1.8.5-r3, 1.9.6-r2, 1.10.1-r1 sys-devel/binutils: 2.18-r1 sys-devel/gcc-config: 1.4.0-r4 sys-devel/libtool: 2.2.4 virtual/os-headers: 2.6.25-r1 ACCEPT_KEYWORDS="x86 ~x86" CBUILD="i686-pc-linux-gnu" CFLAGS="-O2 -march=i686 -pipe" CHOST="i686-pc-linux-gnu" CONFIG_PROTECT="/etc /usr/kde/3.5/env /usr/kde/3.5/share/config /usr/kde/3.5/shutdown /usr/share/config /var/lib/hsqldb" CONFIG_PROTECT_MASK="/etc/env.d /etc/env.d/java/ /etc/fonts/fonts.conf /etc/gconf /etc/gentoo-release /etc/php/apache2-php5/ext-active/ /etc/php/cgi-php5/ext-active/ /etc/php/cli-php5/ext-active/ /etc/revdep-rebuild /etc/sandbox.d /etc/terminfo /etc/texmf/web2c /etc/udev/rules.d" CXXFLAGS="-O2 -march=i686 -pipe" DISTDIR="/usr/portage/distfiles" FEATURES="buildpkg distlocks parallel-fetch sandbox sfperms strict unmerge-orphans userfetch" GENTOO_MIRRORS="ftp://sunsite.informatik.rwth-aachen.de/pub/Linux/gentoo " LDFLAGS="" LINGUAS="en de" MAKEOPTS="-j3" PKGDIR="/usr/portage/packages" PORTAGE_RSYNC_OPTS="--recursive --links --safe-links --perms --times --compress --force --whole-file --delete --stats --timeout=180 --exclude=/distfiles --exclude=/local --exclude=/packages" PORTAGE_TMPDIR="/var/tmp" PORTDIR="/usr/portage" PORTDIR_OVERLAY="/usr/local/portage" SYNC="rsync://rsync.informatik.RWTH-Aachen.de/gentoo-portage" USE="X acl berkdb bitmap-fonts bzip2 cairo cdr cli cracklib crypt cups dbus doc dri dvd fortran gcj gdbm gnome gpm gtk gtk2 hal iconv ipv6 isdnlog jpeg kde midi mudflap ncurses nls nptl nptlonly opengl openmp pam pcre pdf perl png pppd python qt readline reflection session spl sqlite sqlite3 ssl svg tcl tcpd tetex tiff tk truetype-fonts type1-fonts unicode x86 xorg xulrunner zlib" ALSA_CARDS="ali5451 als4000 atiixp atiixp-modem bt87x ca0106 cmipci emu10k1 emu10k1x ens1370 ens1371 es1938 es1968 fm801 hda-intel intel8x0 intel8x0m maestro3 trident usb-audio via82xx via82xx-modem ymfpci" ALSA_PCM_PLUGINS="adpcm alaw asym copy dmix dshare dsnoop empty extplug file hooks iec958 ioplug ladspa lfloat linear meter mulaw multi null plug rate route share shm softvol" APACHE2_MODULES="actions alias auth_basic authn_alias authn_anon authn_dbm authn_default authn_file authz_dbm authz_default authz_groupfile authz_host authz_owner authz_user autoindex cache dav dav_fs dav_lock deflate dir disk_cache env expires ext_filter file_cache filter headers include info log_config logio mem_cache mime mime_magic negotiation rewrite setenvif speling status unique_id userdir usertrack vhost_alias" ELIBC="glibc" INPUT_DEVICES="keyboard mouse" KERNEL="linux" LCD_DEVICES="bayrad cfontz cfontz633 glk hd44780 lb216 lcdm001 mtxorb ncurses text" LINGUAS="en de" USERLAND="GNU" VIDEO_CARDS="nv" Unset: CPPFLAGS, CTARGET, EMERGE_DEFAULT_OPTS, INSTALL_MASK, LANG, LC_ALL, PORTAGE_COMPRESS, PORTAGE_COMPRESS_FLAGS, PORTAGE_RSYNC_EXTRA_OPTS
Here, openafs 1.4.7 is running fine with a _vanilla_ 2.6.25 kernel. (well, i only tested "ls -la /afs" and "find /afs")
On my system too, openafs-1.4.7 works perfectly. Running a 2.6.25-gentoo-r1 kernel on amd64. Anything in "dmesg"? What do you mean by freezing? Does the whole system freeze, or just the tasks that try to access afs? Have you allowed time (i.e. several minutes) to let openafs discover it cannot reach a server? ...
(In reply to comment #2) > On my system too, openafs-1.4.7 works perfectly. Running a 2.6.25-gentoo-r1 > kernel on amd64. > > Anything in "dmesg"? > What do you mean by freezing? Does the whole system freeze, or just the tasks > that try to access afs? Have you allowed time (i.e. several minutes) to let > openafs discover it cannot reach a server? ... No, dmesg shows nothing. Any access in relation to /afs stalls the application, e.g. ls /afs/... df cd /afs/... /etc/init.d/openafs-client stop all hang. And of course no client can access the afs filesystem any more. And of course, a normal shutdown (like init 0 or init 6) doesn't work either since the system cannot unmount /afs . And being forced to do a hardware reset isn't funny. Comments on the openafs mailing list seem to indicate that this is possibly connected to missing time-outs in the code. Having stepped back to kernel 2.6.24-r2 (with openafs-kernel 1.4.6_p20080222) and openafs-1.4.7 there aren't any problems anymore.
I have exactly the same problem. Just after boot everything works, but after a while all AFS related things stop working (ls /afs, bos status, etc) I use gentoo-sources-2.6.25-r2 and opensafs-kernel-1.4.7
I suggest this be handled upstream, in the thread on the openafs-info mailing list that corresponds to this bug. More precisely, I think the question at https://lists.openafs.org/pipermail/openafs-info/2008-May/029267.html could use an answer.
In the light of bug #223355, could you please execute echo 2 > /proc/sys/afs/GCPAGs immediately after starting openafs-client, and check whether the problem persists?
(In reply to comment #6) > In the light of bug #223355, could you please execute > echo 2 > /proc/sys/afs/GCPAGs > immediately after starting openafs-client, and check whether the problem > persists? > Initial experience is excellent! I've put this line into /etc/init.d/openafs-client
I have a hunch this is fixed in openafs-1.4.8. Could you verify that this works for you?
As there is no response from the reporter, assuming the answer to "does it work for you in 1.4.8?" is "yes". Closing as fixed.
