[Dovecot] dovecot: Fatal: Time just moved backwards by 3909 seconds.
Hi,
after setting up a KVM based virtual guest on one of our virtualization servers, we see dovecot die on that virtual guest regularly because of an alleged time shift.
The virtual guest is Centos 5.4 based with dovecot 1.2.8 (at first we also tried with the original 1.0.7 (?) dovecot shipped with Centos).
I wrote "alleged time shift" because there is no timeshift whatsoever, or at least I don't notice it anywhere else.
ntp is - of course ;-) - up and running and no other applications have time troubles and I've even disabled the local clock as a time reference.
As a proof that the time did not shift see this log excerpt:
------CUT------- Nov 29 14:02:22 athesia dovecot: IMAP(atester): Connection closed Nov 29 14:04:11 athesia dovecot: imap-login: Time just moved backwards by 4397 seconds. This might cause a lot of problems, so I'll just kill myself now. http://wiki.dovecot.org/TimeMovedBackwards ------CUT-------
So as you see, dovecot successfully logged a message on 14:02 and then complained about the time having shifted 4397 seconds on 14:04 ...
Thanks in advance
-- Udo Rader, CTO http://www.bestsolution.at http://riaschissl.blogspot.com
On 11/30/2009, Udo Rader (listudo@bestsolution.at) wrote:
The virtual guest is Centos 5.4 based with dovecot 1.2.8 (at first we also tried with the original 1.0.7 (?) dovecot shipped with Centos).
I wrote "alleged time shift" because there is no timeshift whatsoever, or at least I don't notice it anywhere else.
ntp is - of course ;-) - up and running and no other applications have time troubles and I've even disabled the local clock as a time reference.
Of course you know not to use ntp on VMs, only on the host, right?
:)
Revent kernels should be able to keep the VM time synced using kvmclock clocksource...
Hello, I'm experimenting with the Lazyexpunge plugin on a test machine with Dovecot 1.2.8 on Debian 5.0.
This is my conf
# 1.2.8: /usr/local/etc/dovecot.conf # OS: Linux 2.6.26-2-xen-686 i686 Debian 5.0.3 log_timestamp: %Y-%m-%d %H:%M:%S protocols: imap pop3 imaps pop3s ssl_cert_file: /usr/local/etc/dovecot.crt ssl_key_file: /usr/local/etc/dovecot.key disable_plaintext_auth: no login_dir: /usr/local/var/run/dovecot/login login_executable(default): /usr/local/libexec/dovecot/imap-login login_executable(imap): /usr/local/libexec/dovecot/imap-login login_executable(pop3): /usr/local/libexec/dovecot/pop3-login login_greeting: Server ready. max_mail_processes: 256 mail_max_userip_connections(default): 10 mail_max_userip_connections(imap): 10 mail_max_userip_connections(pop3): 3 first_valid_uid: 100 last_valid_uid: 105 mail_privileged_group: Debian-exim mail_uid: Debian-exim mail_gid: Debian-exim mail_location: maildir:~/Maildir:INDEX=/var/indexes/%d/%n mail_debug: yes fsync_disable: yes mail_nfs_storage: no lock_method: dotlock mail_drop_priv_before_exec: yes mail_executable(default): /usr/local/libexec/dovecot/imap mail_executable(imap): /usr/local/libexec/dovecot/imap mail_executable(pop3): /usr/local/libexec/dovecot/pop3 mail_plugins(default): quota imap_quota mail_log lazy_expunge mail_plugins(imap): quota imap_quota mail_log lazy_expunge mail_plugins(pop3): quota mail_log mail_plugin_dir(default): /usr/local/lib/dovecot/imap mail_plugin_dir(imap): /usr/local/lib/dovecot/imap mail_plugin_dir(pop3): /usr/local/lib/dovecot/pop3 pop3_no_flag_updates: yes namespace: type: private separator: / inbox: yes list: yes subscriptions: yes namespace: type: private separator: / prefix: EXPUNGED/ location: maildir:~/Maildir/expunged list: yes subscriptions: yes namespace: type: private separator: / prefix: DELETED/ location: maildir:~/Maildir/deleted list: yes subscriptions: yes namespace: type: private separator: / prefix: DELETED/EXPUNGED/ location: maildir:~/Maildir/deleted/expunged list: yes subscriptions: yes lda: mail_plugins: quota sieve postmaster_address: postmaster@securitylabs.it auth default: cache_size: 1000 cache_ttl: 120 cache_negative_ttl: 120 user: Debian-exim username_chars: abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ01234567890.-_@& master_user_separator: * passdb: driver: sql args: /usr/local/etc/dovecot-sql.conf passdb: driver: passwd-file args: /usr/local/etc/passwd.masterusers pass: yes master: yes userdb: driver: sql args: /usr/local/etc/dovecot-sql.conf socket: type: listen master: path: /usr/local/var/run/dovecot/auth-master mode: 384 user: 102 group: 104 plugin: quota: maildir quota_warning: storage=80%% /usr/local/bin/dovecot-quota-warning.sh 80 quota_warning2: storage=90%% /usr/local/bin/dovecot-quota-warning.sh 90 mail_log_events: delete expunge mail_log_group_events: sieve: ~/.dovecot.sieve lazy_expunge: EXPUNGED/ DELETED/ DELETED/EXPUNGED/
I've added the namespace and enabled the plugin for the test as reported here:
http://wiki.dovecot.org/Plugins/Lazyexpunge.
Now every time I try to expunge a mailbox dovecot crash. Here the strace:
15:09:21.054191 gettimeofday({1259590161, 54371}, {0, 0}) = 0 15:09:21.054418 gettimeofday({1259590161, 54438}, NULL) = 0 15:09:21.054530 epoll_wait(5, {}, 4, 3159) = 0 15:09:24.213497 gettimeofday({1259590164, 213531}, {0, 0}) = 0 15:09:24.213574 gettimeofday({1259590164, 213591}, NULL) = 0 15:09:24.213623 epoll_wait(5, {{EPOLLIN, {u32=146866728, u64=146866728}}}, 4, 3207) = 1 15:09:25.769640 gettimeofday({1259590165, 769672}, {0, 0}) = 0 15:09:25.769709 gettimeofday({1259590165, 769726}, NULL) = 0 15:09:25.769760 read(0, "DONE\r\n"..., 4015) = 6 15:09:25.769836 epoll_ctl(5, EPOLL_CTL_DEL, 0, {0, {u32=146866728, u64=146866728}}) = 0 15:09:25.769895 inotify_rm_watch(10, 2) = 0 15:09:25.770188 inotify_rm_watch(10, 1) = 0 15:09:25.770236 epoll_ctl(5, EPOLL_CTL_DEL, 10, {0, {u32=147113264, u64=147113264}}) = 0 15:09:25.770295 setsockopt(1, SOL_TCP, TCP_CORK, [1], 4) = 0 15:09:25.770353 write(1, "5 OK Idle completed.\r\n"..., 22) = 22 15:09:25.770416 setsockopt(1, SOL_TCP, TCP_CORK, [0], 4) = 0 15:09:25.770477 gettimeofday({1259590165, 770504}, NULL) = 0 15:09:25.770542 epoll_ctl(5, EPOLL_CTL_ADD, 0, {EPOLLIN|EPOLLPRI|EPOLLERR|EPOLLHUP, {u32=146866728, u64=146866728}}) = 0 15:09:25.770601 gettimeofday({1259590165, 770649}, NULL) = 0 15:09:25.770681 epoll_wait(5, {{EPOLLIN, {u32=146866728, u64=146866728}}}, 4, 3443) = 1 15:09:25.804121 gettimeofday({1259590165, 804143}, {0, 0}) = 0 15:09:25.804177 gettimeofday({1259590165, 804193}, NULL) = 0 15:09:25.804224 read(0, "6 uid store 10162:10163,10165:101"..., 4009) = 61 15:09:25.804324 gettimeofday({1259590165, 804343}, NULL) = 0 15:09:25.804405 stat64("/var/indexes/securitylabs.it/mailing/.Debian/dovecot.index.log", {st_mode=S_IFREG|0600, st_size=5684, ...}) = 0 15:09:25.804539 fstat64(6, {st_mode=S_IFREG|0600, st_size=5684, ...}) = 0 15:09:25.804610 fstat64(6, {st_mode=S_IFREG|0600, st_size=5684, ...}) = 0 15:09:25.804697 stat64("/var/vexim/securitylabs.it/mailing/Maildir/.Debian/new", {st_mode=S_IFDIR|0700, st_size=40960, ...}) = 0 15:09:25.804790 stat64("/var/vexim/securitylabs.it/mailing/Maildir/.Debian/cur", {st_mode=S_IFDIR|0700, st_size=581632, ...}) = 0 15:09:25.804880 stat64("/var/vexim/securitylabs.it/mailing/Maildir/.Debian", {st_mode=S_IFDIR|0700, st_size=4096, ...}) = 0 15:09:25.804967 chown32("/var/vexim/securitylabs.it/mailing/Maildir/.Debian", 102, -1) = 0 15:09:25.805033 open("/var/vexim/securitylabs.it/mailing/Maildir/.Debian/dovecot-uidlist", O_RDONLY|O_LARGEFILE) = 11 15:09:25.805103 close(11) = 0 15:09:25.805148 stat64("/var/vexim/securitylabs.it/mailing/Maildir/.Debian/dovecot-uidlist", {st_mode=S_IFREG|0600, st_size=66630, ...}) = 0 15:09:25.805239 fstat64(8, {st_mode=S_IFREG|0600, st_size=66630, ...}) = 0 15:09:25.805311 stat64("/var/indexes/securitylabs.it/mailing/.Debian/dovecot.index.log", {st_mode=S_IFREG|0600, st_size=5684, ...}) = 0 15:09:25.805402 fstat64(6, {st_mode=S_IFREG|0600, st_size=5684, ...}) = 0 15:09:25.805470 fstat64(6, {st_mode=S_IFREG|0600, st_size=5684, ...}) = 0 15:09:25.805570 stat64("/var/indexes/securitylabs.it/mailing/.Debian/dovecot.index.log", {st_mode=S_IFREG|0600, st_size=5684, ...}) = 0 15:09:25.805664 fstat64(6, {st_mode=S_IFREG|0600, st_size=5684, ...}) = 0 15:09:25.805732 fstat64(6, {st_mode=S_IFREG|0600, st_size=5684, ...}) = 0 15:09:25.805809 gettimeofday({1259590165, 805826}, NULL) = 0 15:09:25.805864 setsockopt(1, SOL_TCP, TCP_CORK, [1], 4) = 0 15:09:25.805912 write(1, "6 OK Store completed.\r\n"..., 23) = 23 15:09:25.805973 setsockopt(1, SOL_TCP, TCP_CORK, [0], 4) = 0 15:09:25.806032 gettimeofday({1259590165, 806057}, NULL) = 0 15:09:25.806089 epoll_wait(5, {{EPOLLIN, {u32=146866728, u64=146866728}}}, 4, 3407) = 1 15:09:25.816436 gettimeofday({1259590165, 816458}, {0, 0}) = 0 15:09:25.816491 gettimeofday({1259590165, 816508}, NULL) = 0 15:09:25.816539 read(0, "7 expunge\r\n"..., 3948) = 11 15:09:25.816608 stat64("/var/indexes/securitylabs.it/mailing/.Debian/dovecot.index.log", {st_mode=S_IFREG|0600, st_size=5684, ...}) = 0 15:09:25.816706 fstat64(6, {st_mode=S_IFREG|0600, st_size=5684, ...}) = 0 15:09:25.816775 fstat64(6, {st_mode=S_IFREG|0600, st_size=5684, ...}) = 0 15:09:25.816889 gettimeofday({1259590165, 816908}, NULL) = 0 15:09:25.816954 --- SIGSEGV (Segmentation fault) @ 0 (0) ---
Another trouble I have is that now every time i select (with Thunderbird) a special folder (like "Sent", "Drafts") I receive this error:
"The current command did not succeed. Tha mail server responded:Character not allowed in mailbox name: '.' "
I've never configured namespace before, so I suspect a misconfiguration but I can't figure out.
P.S. this machine is behind a proxy with same version of dovecot and same OS. I don't have modified the configuration of the proxy.
Thanks for the support, Igor
On Nov 30, 2009, at 9:25 AM, mailing@securitylabs.it wrote:
Now every time I try to expunge a mailbox dovecot crash. Here the strace:
strace is almost useless for figuring out why program crashes. Could you get gdb backtrace? See http://dovecot.org/bugreport.html
Another trouble I have is that now every time i select (with Thunderbird) a special folder (like "Sent", "Drafts") I receive this error:
"The current command did not succeed. Tha mail server responded:Character not allowed in mailbox name: '.' "
I've never configured namespace before, so I suspect a misconfiguration but I can't figure out.
Did you migrate to Dovecot from another server? Sounds anyway like TB is configured to have "INBOX." namespace prefix or something like that. Or possibly it's caching wrong hierarchy separator and you need to recreate the account in TB.
Timo Sirainen ha scritto:
On Nov 30, 2009, at 9:25 AM, mailing@securitylabs.it wrote:
Now every time I try to expunge a mailbox dovecot crash. Here the strace:
strace is almost useless for figuring out why program crashes. Could you get gdb backtrace? See http://dovecot.org/bugreport.html
Hello, Timo here gdb backtrace:
GNU gdb 6.8-debian Copyright (C) 2008 Free Software Foundation, Inc. License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html> This is free software: you are free to change and redistribute it. There is NO WARRANTY, to the extent permitted by law. Type "show copying" and "show warranty" for details. This GDB was configured as "i486-linux-gnu"...
warning: Can't read pathname for load map: Input/output error. Reading symbols from /lib/i686/nosegneg/libdl.so.2...done. Loaded symbols for /lib/i686/nosegneg/libdl.so.2 Reading symbols from /lib/i686/nosegneg/librt.so.1...done. Loaded symbols for /lib/i686/nosegneg/librt.so.1 Reading symbols from /lib/i686/nosegneg/libc.so.6...done. Loaded symbols for /lib/i686/nosegneg/libc.so.6 Reading symbols from /lib/ld-linux.so.2...done. Loaded symbols for /lib/ld-linux.so.2 Reading symbols from /lib/i686/nosegneg/libpthread.so.0...done. Loaded symbols for /lib/i686/nosegneg/libpthread.so.0 Reading symbols from /usr/local/lib/dovecot/imap/lib02_lazy_expunge_plugin.so...done. Loaded symbols for /usr/local/lib/dovecot/imap/lib02_lazy_expunge_plugin.so Reading symbols from /usr/local/lib/dovecot/imap/lib10_quota_plugin.so...done. Loaded symbols for /usr/local/lib/dovecot/imap/lib10_quota_plugin.so Reading symbols from /usr/local/lib/dovecot/imap/lib11_imap_quota_plugin.so...done. Loaded symbols for /usr/local/lib/dovecot/imap/lib11_imap_quota_plugin.so Reading symbols from /usr/local/lib/dovecot/imap/lib20_mail_log_plugin.so...done. Loaded symbols for /usr/local/lib/dovecot/imap/lib20_mail_log_plugin.so Core was generated by `imap'. Program terminated with signal 11, Segmentation fault. [New process 32041] #0 lazy_expunge_mail_expunge (_mail=0x9907ae8) at lazy-expunge-plugin.c:116 116 lt->expunge_box = mailbox_open_or_create(deststorage, (gdb) bt full #0 lazy_expunge_mail_expunge (_mail=0x9907ae8) at lazy-expunge-plugin.c:116 lt = (struct lazy_expunge_transaction *) 0x99079b8 #1 0x08069960 in imap_expunge (box=0x98db538, next_search_arg=0x0) at imap-expunge.c:35 ctx = (struct mail_search_context *) 0x99079d0 t = (struct mailbox_transaction_context *) 0x98fea28 mail = (struct mail *) 0x9907ae8 search_args = (struct mail_search_args *) 0x0 expunges = true #2 0x08061dc8 in cmd_expunge_finish (cmd=0x98d64d0, search_args=0x0) at cmd-expunge.c:27 client = (struct client *) 0x98d6250 #3 0x08067b4c in client_command_input (cmd=0x98d64d0) at client.c:612 client = (struct client *) 0x98d6250 command = <value optimized out> __PRETTY_FUNCTION__ = "client_command_input" #4 0x08067be9 in client_command_input (cmd=0x98d64d0) at client.c:661 client = (struct client *) 0x98d6250 command = <value optimized out> __PRETTY_FUNCTION__ = "client_command_input" #5 0x08067d5d in client_handle_input (client=0x98d6250) at client.c:702 _data_stack_cur_id = 3 ret = false remove_io = <value optimized out> handled_commands = false __PRETTY_FUNCTION__ = "client_handle_input" #6 0x080686bf in client_input (client=0x98d6250) at client.c:753 cmd = <value optimized out> output = (struct ostream *) 0x98d6404 bytes = <value optimized out> __PRETTY_FUNCTION__ = "client_input" #7 0x080fb8c0 in io_loop_handler_run (ioloop=0x98d19b0) at ioloop-epoll.c:208 ctx = (struct ioloop_handler_context *) 0x98d1ab8 event = (const struct epoll_event *) 0x98d1af8 list = (struct io_list *) 0x98d4228 io = (struct io_file *) 0x98d6460 tv = {tv_sec = 1799, tv_usec = 999278} t_id = 2 msecs = <value optimized out> ret = 1 i = 0 j = 0 call = <value optimized out> #8 0x080fad30 in io_loop_run (ioloop=0x98d19b0) at ioloop.c:335 No locals. #9 0x0807119a in main (argc=Cannot access memory at address 0xc ) at main.c:327 No locals.
Did you migrate to Dovecot from another server? Sounds anyway like TB is configured to have "INBOX." namespace prefix or something like that. Or possibly it's caching wrong hierarchy separator and you need to recreate the account in TB.
With a different version of TB on another computer this does not happen, tomorrow I will try to recreate the account on the problematic TB.
Thanks, Igor
On Mon, 2009-11-30 at 19:52 +0100, mailing@securitylabs.it wrote:
strace is almost useless for figuring out why program crashes. Could you get gdb backtrace? See http://dovecot.org/bugreport.html
Hello, Timo here gdb backtrace:
You still happen to have the core file? I'd like to know a few more things:
#0 lazy_expunge_mail_expunge (_mail=0x9907ae8) at lazy-expunge-plugin.c:116 116 lt->expunge_box =
p *lt p *deststorage p *_mail p *_mail.box
Anyway there's something weird going on there. It shouldn't crash on that line. So either gcc optimizations confused gdb and it's actually crashing elsewhere (recompiling+reinstalling the plugin without -O2 parameter would help with this), or there is some memory corruption which is probably going to be tricky to find without valgrind.
On 09/12/2009 21:03, Timo Sirainen wrote:
You still happen to have the core file? I'd like to know a few more things:
#0 lazy_expunge_mail_expunge (_mail=0x9907ae8) at lazy-expunge-plugin.c:116 116 lt->expunge_box =
p *lt p *deststorage p *_mail p *_mail.box
Anyway there's something weird going on there. It shouldn't crash on that line. So either gcc optimizations confused gdb and it's actually crashing elsewhere (recompiling+reinstalling the plugin without -O2 parameter would help with this),
Hello, I've recompiled dovecot with -O0 and here a new backtrace:
GNU gdb 6.8-debian Copyright (C) 2008 Free Software Foundation, Inc. License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html> This is free software: you are free to change and redistribute it. There is NO WARRANTY, to the extent permitted by law. Type "show copying" and "show warranty" for details. This GDB was configured as "i486-linux-gnu"...
warning: Can't read pathname for load map: Input/output error. Reading symbols from /lib/i686/nosegneg/libdl.so.2...done. Loaded symbols for /lib/i686/nosegneg/libdl.so.2 Reading symbols from /lib/i686/nosegneg/librt.so.1...done. Loaded symbols for /lib/i686/nosegneg/librt.so.1 Reading symbols from /lib/i686/nosegneg/libc.so.6...done. Loaded symbols for /lib/i686/nosegneg/libc.so.6 Reading symbols from /lib/ld-linux.so.2...done. Loaded symbols for /lib/ld-linux.so.2 Reading symbols from /lib/i686/nosegneg/libpthread.so.0...done. Loaded symbols for /lib/i686/nosegneg/libpthread.so.0 Reading symbols from /usr/local/lib/dovecot/imap/lib02_lazy_expunge_plugin.so...done. Loaded symbols for /usr/local/lib/dovecot/imap/lib02_lazy_expunge_plugin.so Reading symbols from /usr/local/lib/dovecot/imap/lib10_quota_plugin.so...done. Loaded symbols for /usr/local/lib/dovecot/imap/lib10_quota_plugin.so Reading symbols from /usr/local/lib/dovecot/imap/lib11_imap_quota_plugin.so...done. Loaded symbols for /usr/local/lib/dovecot/imap/lib11_imap_quota_plugin.so Reading symbols from /usr/local/lib/dovecot/imap/lib20_mail_log_plugin.so...done. Loaded symbols for /usr/local/lib/dovecot/imap/lib20_mail_log_plugin.so Core was generated by `imap'. Program terminated with signal 11, Segmentation fault. [New process 14769] #0 0xb7fb45c9 in lazy_expunge_mail_expunge (_mail=0x844df58) at lazy-expunge-plugin.c:115 115 deststorage = luser->lazy_ns[LAZY_NAMESPACE_EXPUNGE]->storage; (gdb) bt full #0 0xb7fb45c9 in lazy_expunge_mail_expunge (_mail=0x844df58) at lazy-expunge-plugin.c:115 luser = (struct lazy_expunge_mail_user *) 0x841c060 lt = (struct lazy_expunge_transaction *) 0x844de28 deststorage = (struct mail_storage *) 0x844de40 #1 0x080c658f in mail_expunge (mail=0x844df58) at mail.c:207 p = (struct mail_private *) 0x844df58 #2 0x0806c294 in imap_expunge (box=0x8423538, next_search_arg=0x0) at imap-expunge.c:35 ctx = (struct mail_search_context *) 0x844de40 t = (struct mailbox_transaction_context *) 0x844ca58 mail = (struct mail *) 0x844df58 search_args = (struct mail_search_args *) 0x0 expunges = false #3 0x0806267e in cmd_expunge_finish (cmd=0x841e4d0, search_args=0x0) at cmd-expunge.c:27 client = (struct client *) 0x841e250 #4 0x0806283f in cmd_expunge (cmd=0x841e4d0) at cmd-expunge.c:78 No locals. #5 0x0806a3c7 in client_command_input (cmd=0x841e4d0) at client.c:612 client = (struct client *) 0x841e250 command = (struct command *) 0x2 __PRETTY_FUNCTION__ = "client_command_input" #6 0x0806a5f3 in client_command_input (cmd=0x841e4d0) at client.c:661 client = (struct client *) 0x841e250 command = (struct command *) 0x841a6a0 __PRETTY_FUNCTION__ = "client_command_input" #7 0x0806a6fe in client_handle_next_command (client=0x841e250, remove_io_r=0xbfb94035) at client.c:702 size = 11 #8 0x0806a783 in client_handle_input (client=0x841e250) at client.c:714 _data_stack_cur_id = 3 ret = 65 remove_io = false handled_commands = false __PRETTY_FUNCTION__ = "client_handle_input" #9 0x0806a8e1 in client_input (client=0x841e250) at client.c:753 cmd = (struct client_command_context *) 0xb7f2af78 output = (struct ostream *) 0x841e404 bytes = 11 __PRETTY_FUNCTION__ = "client_input" #10 0x08124721 in io_loop_handler_run (ioloop=0x84199b0) at ioloop-epoll.c:208 ctx = (struct ioloop_handler_context *) 0x8419ab8 events = (struct epoll_event *) 0x8419af8 event = (const struct epoll_event *) 0x8419af8 list = (struct io_list *) 0x841c228 io = (struct io_file *) 0x841e460 tv = {tv_sec = 1799, tv_usec = 999144} events_count = 3 t_id = 2 msecs = 1800000 ret = 1 i = 0 j = 0 call = true #11 0x081239d8 in io_loop_run (ioloop=0x84199b0) at ioloop.c:335 No locals. #12 0x08075b16 in main (argc=1, argv=0xbfb94194, envp=0xbfb9419c) at main.c:327 No locals.
or there is some memory corruption which is probably going to be tricky to find without valgrind.
That's a virtual machine inside Xen, may be a problem?
Thanks for your support, Igor
On Thu, 2009-12-10 at 11:55 +0100, mailing@securitylabs.it wrote:
#0 0xb7fb45c9 in lazy_expunge_mail_expunge (_mail=0x844df58) at lazy-expunge-plugin.c:115 115 deststorage = luser->lazy_ns[LAZY_NAMESPACE_EXPUNGE]->storage;
OK, this makes more sense. Looks like it was a bug in quota plugin: http://hg.dovecot.org/dovecot-1.2/rev/5e2206e3c750
Il 10/12/2009 19:28, Timo Sirainen ha scritto:
On Thu, 2009-12-10 at 11:55 +0100, mailing@securitylabs.it wrote:
#0 0xb7fb45c9 in lazy_expunge_mail_expunge (_mail=0x844df58) at lazy-expunge-plugin.c:115 115 deststorage = luser->lazy_ns[LAZY_NAMESPACE_EXPUNGE]->storage;
OK, this makes more sense. Looks like it was a bug in quota plugin: http://hg.dovecot.org/dovecot-1.2/rev/5e2206e3c750
Hello Timo, thanks problem solved. Igor.
Charles Marcus wrote:
On 11/30/2009, Udo Rader (listudo@bestsolution.at) wrote:
The virtual guest is Centos 5.4 based with dovecot 1.2.8 (at first we also tried with the original 1.0.7 (?) dovecot shipped with Centos).
I wrote "alleged time shift" because there is no timeshift whatsoever, or at least I don't notice it anywhere else.
ntp is - of course ;-) - up and running and no other applications have time troubles and I've even disabled the local clock as a time reference.
Of course you know not to use ntp on VMs, only on the host, right?
:)
heh, ok, ship hit an sunk :-)
I was absolutely not aware of a "clocksource" kernel parameter, what a weird thing ...
Revent kernels should be able to keep the VM time synced using kvmclock clocksource...
So I'll give clocksource=acpi_pm a chance and see how it turns out ...
Thanks for your hint!
-- Udo Rader, CTO http://www.bestsolution.at http://riaschissl.blogspot.com
Udo Rader wrote:
Charles Marcus wrote:
On 11/30/2009, Udo Rader (listudo@bestsolution.at) wrote:
The virtual guest is Centos 5.4 based with dovecot 1.2.8 (at first we also tried with the original 1.0.7 (?) dovecot shipped with Centos).
I wrote "alleged time shift" because there is no timeshift whatsoever, or at least I don't notice it anywhere else.
ntp is - of course ;-) - up and running and no other applications have time troubles and I've even disabled the local clock as a time reference.
Of course you know not to use ntp on VMs, only on the host, right?
:)
heh, ok, ship hit an sunk :-)
I was absolutely not aware of a "clocksource" kernel parameter, what a weird thing ...
Revent kernels should be able to keep the VM time synced using kvmclock clocksource...
So I'll give clocksource=acpi_pm a chance and see how it turns out ...
So for the sake of other peoples' nerves also facing this problem, the solution was to add "divider=10" as a kernel boot parameter.
There is an good post about this problem here:
http://forums.fedoraforum.org/showthread.php?t=211100
and this also leads to this document for VMWare giving an indication of what the suggested parameters for various distributions are:
http://kb.vmware.com/kb/1006427
cheers!
-- Udo Rader, CTO http://www.bestsolution.at http://riaschissl.blogspot.com
On 11/30/2009 4:36 PM, Udo Rader wrote:
Recent kernels should be able to keep the VM time synced using kvmclock clocksource...
So I'll give clocksource=acpi_pm a chance and see how it turns out ...
So for the sake of other peoples' nerves also facing this problem, the solution was to add "divider=10" as a kernel boot parameter.
There is an good post about this problem here:
http://forums.fedoraforum.org/showthread.php?t=211100
and this also leads to this document for VMWare giving an indication of what the suggested parameters for various distributions are:
Did you not try the clocksource=kvm-clock? Or do you not have a new enough kernel (2.6.27+)?
Charles Marcus wrote:
On 11/30/2009 4:36 PM, Udo Rader wrote:
Recent kernels should be able to keep the VM time synced using kvmclock clocksource...
So I'll give clocksource=acpi_pm a chance and see how it turns out ...
So for the sake of other peoples' nerves also facing this problem, the solution was to add "divider=10" as a kernel boot parameter.
There is an good post about this problem here:
http://forums.fedoraforum.org/showthread.php?t=211100
and this also leads to this document for VMWare giving an indication of what the suggested parameters for various distributions are:
Did you not try the clocksource=kvm-clock? Or do you not have a new enough kernel (2.6.27+)?
No, unfortunately Centos 5.4 comes with a 2.6.18 kernel (though with many backports from newer versions).
% uname -r 2.6.18-164.6.1.el5.centos.plusPAE % cat /sys/devices/system/clocksource/clocksource0/current_clocksource kvm-clock
-- Udo Rader, CTO http://www.bestsolution.at http://riaschissl.blogspot.com
participants (4)
-
Charles Marcus
-
mailing＠securitylabs.it
-
Timo Sirainen
-
Udo Rader