Hello, On 8/4/21 9:01 PM, Oswald Buddenhagen wrote:
On Wed, Aug 04, 2021 at 06:20:33PM +0200, Uwe Kleine-König wrote:ok, that's "interesting": the job not only takes lots of wall time, it also consumes huge amounts of cpu time. something is seriously fishy in that setup. first to check would be whether it's ssh or mbsync (and if it's the latter, whether it's openssl or something else).A bigger context of the logs of an older run can be found at https://www.kleine-koenig.org/~uwe/syncmail.log
It currently happens again :-\ strace looks as follows: ...poll([{fd=4, events=POLLIN|POLLOUT}], 1, 2000) = 1 ([{fd=4, revents=POLLIN}]) write(4, "\27\3\3\0F\250\304\334\237W\235\257\307\276\256fiy\374[\3057\"V\25\332\370V\366<Q\321"..., 75) = -1 EAGAIN (Resource temporarily unavailable)
read(4, "\27\3\3\0\365", 5) = 5read(4, "%{&\27\326UVl\2657F\207\331\305i\356\200ot\37\36se\312i_\272\310\0\33\352\211"..., 245) = 245
lseek(3, 0, SEEK_SET) = 0 write(3, "1591221302\n4727926\n", 19) = 19 ftruncate(3, 19) = 0 fdatasync(3) = 0openat(AT_FDCWD, "/home/uwe/Maildir-work/.lists.linux-arm-kernel/tmp/1631042587.213028_341577.taurus,U=4727926:2,", O_WRONLY|O_CREAT|O_EXCL, 0600) = 8
write(8, "Return-path: <linux-arm-kernel-b"..., 7718) = 7718 fsync(8) = 0 close(8) = 0 utime("/home/uwe/Maildir-work/.lists.linux-arm-kernel/tmp/1631042587.213028_341577.taurus,U=4727926:2,", {actime=1562768095 /* 2019-07-10T16:14:55+0200 */, modtime=1562768095 /* 2019-07-10T16:14:55+0200 */}) = 0 rename("/home/uwe/Maildir-work/.lists.linux-arm-kernel/tmp/1631042587.213028_341577.taurus,U=4727926:2,", "/home/uwe/Maildir-work/.lists.linux-arm-kernel/new/1631042587.213028_341577.taurus,U=4727926:2,") = 0 write(7, "> 4728198 0 4727926\n", 20) = 20 lseek(3, 0, SEEK_SET) = 0 write(3, "1591221302\n4727927\n", 19) = 19 ftruncate(3, 19) = 0 fdatasync(3) = 0openat(AT_FDCWD, "/home/uwe/Maildir-work/.lists.linux-arm-kernel/tmp/1631042587.213028_341578.taurus,U=4727927:2,", O_WRONLY|O_CREAT|O_EXCL, 0600) = 8
write(8, "Return-path: <linux-arm-kernel-b"..., 7409) = 7409 fsync(8) = 0 close(8) = 0 utime("/home/uwe/Maildir-work/.lists.linux-arm-kernel/tmp/1631042587.213028_341578.taurus,U=4727927:2,", {actime=1562661698 /* 2019-07-09T10:41:38+0200 */, modtime=1562661698 /* 2019-07-09T10:41:38+0200 */}) = 0 rename("/home/uwe/Maildir-work/.lists.linux-arm-kernel/tmp/1631042587.213028_341578.taurus,U=4727927:2,", "/home/uwe/Maildir-work/.lists.linux-arm-kernel/new/1631042587.213028_341578.taurus,U=4727927:2,") = 0 write(7, "> 4728199 0 4727927\n", 20) = 20 lseek(3, 0, SEEK_SET) = 0 write(3, "1591221302\n4727928\n", 19) = 19 ftruncate(3, 19) = 0 fdatasync(3) = 0openat(AT_FDCWD, "/home/uwe/Maildir-work/.lists.linux-arm-kernel/tmp/1631042587.213028_341579.taurus,U=4727928:2,", O_WRONLY|O_CREAT|O_EXCL, 0600) = 8
write(8, "Return-path: <linux-arm-kernel-b"..., 10959) = 10959 fsync(8) = 0 close(8) = 0 utime("/home/uwe/Maildir-work/.lists.linux-arm-kernel/tmp/1631042587.213028_341579.taurus,U=4727928:2,", {actime=1562341587 /* 2019-07-05T17:46:27+0200 */, modtime=1562341587 /* 2019-07-05T17:46:27+0200 */}) = 0 rename("/home/uwe/Maildir-work/.lists.linux-arm-kernel/tmp/1631042587.213028_341579.taurus,U=4727928:2,", "/home/uwe/Maildir-work/.lists.linux-arm-kernel/new/1631042587.213028_341579.taurus,U=4727928:2,") = 0 write(7, "> 4728200 0 4727928\n", 20) = 20 lseek(3, 0, SEEK_SET) = 0 write(3, "1591221302\n4727929\n", 19) = 19 ftruncate(3, 19) = 0 fdatasync(3) = 0openat(AT_FDCWD, "/home/uwe/Maildir-work/.lists.linux-arm-kernel/tmp/1631042587.213028_341580.taurus,U=4727929:2,", O_WRONLY|O_CREAT|O_EXCL, 0600) = 8
write(8, "Return-path: <linux-arm-kernel-b"..., 10563) = 10563 fsync(8) = 0 close(8) = 0 utime("/home/uwe/Maildir-work/.lists.linux-arm-kernel/tmp/1631042587.213028_341580.taurus,U=4727929:2,", {actime=1562736076 /* 2019-07-10T07:21:16+0200 */, modtime=1562736076 /* 2019-07-10T07:21:16+0200 */}) = 0 rename("/home/uwe/Maildir-work/.lists.linux-arm-kernel/tmp/1631042587.213028_341580.taurus,U=4727929:2,", "/home/uwe/Maildir-work/.lists.linux-arm-kernel/new/1631042587.213028_341580.taurus,U=4727929:2,") = 0 write(7, "> 4728201 0 4727929\n", 20) = 20poll([{fd=4, events=POLLIN|POLLOUT}], 1, 2000) = 1 ([{fd=4, revents=POLLIN}]) write(4, "\27\3\3\0F\250\304\334\237W\235\257\307\276\256fiy\374[\3057\"V\25\332\370V\366<Q\321"..., 75) = -1 EAGAIN (Resource temporarily unavailable)
read(4, "\27\3\3\20\21", 5) = 5read(4, "\276\352\253\355\3733\237vz\340\177\330\254\307\366\356\243\2355em\323\272*\312\24\247\276\333\0\334\216"..., 4113) = 4113 poll([{fd=4, events=POLLIN|POLLOUT}], 1, 2000) = 1 ([{fd=4, revents=POLLIN}]) write(4, "\27\3\3\0F\250\304\334\237W\235\257\307\276\256fiy\374[\3057\"V\25\332\370V\366<Q\321"..., 75) = -1 EAGAIN (Resource temporarily unavailable)
read(4, "\27\3\3\2J", 5) = 5read(4, "Qq\17\341\271\205\353\n\210\236\371H\340\235\r\241)S\251p\257\205\36Z\245K\30\222\347\346\224 "..., 586) = 586 poll([{fd=4, events=POLLIN|POLLOUT}], 1, 2000) = 1 ([{fd=4, revents=POLLIN}]) write(4, "\27\3\3\0F\250\304\334\237W\235\257\307\276\256fiy\374[\3057\"V\25\332\370V\366<Q\321"..., 75) = -1 EAGAIN (Resource temporarily unavailable)
read(4, "\27\3\3\vX", 5) = 5read(4, "\320\32\217HM\226y\362\17\5\320\363\2\35z\365\320Bl\6+\305dZ^\371<\373\200zU\334"..., 2904) = 2904 poll([{fd=4, events=POLLIN|POLLOUT}], 1, 2000) = 1 ([{fd=4, revents=POLLIN}]) write(4, "\27\3\3\0F\250\304\334\237W\235\257\307\276\256fiy\374[\3057\"V\25\332\370V\366<Q\321"..., 75) = -1 EAGAIN (Resource temporarily unavailable)
read(4, "\27\3\3\20\21", 5) = 5read(4, "\225\363\341W\213\327\345$n&d\2\27m\265\rR`\260D\356g4ERz\31330\305\323\263"..., 4113) = 4113 poll([{fd=4, events=POLLIN|POLLOUT}], 1, 2000) = 1 ([{fd=4, revents=POLLIN}]) write(4, "\27\3\3\0F\250\304\334\237W\235\257\307\276\256fiy\374[\3057\"V\25\332\370V\366<Q\321"..., 75) = -1 EAGAIN (Resource temporarily unavailable)
read(4, "\27\3\3\4\312", 5) = 5read(4, "\26\31D_\21Y\201\17\34\373\274\263[]\260\245\243\351*u\235\376\2633/\300\225w\304\274\16U"..., 1226) = 1226 poll([{fd=4, events=POLLIN|POLLOUT}], 1, 2000) = 1 ([{fd=4, revents=POLLIN}]) write(4, "\27\3\3\0F\250\304\334\237W\235\257\307\276\256fiy\374[\3057\"V\25\332\370V\366<Q\321"..., 75) = -1 EAGAIN (Resource temporarily unavailable)
read(4, "\27\3\3\t\220", 5) = 5read(4, "CG\272s%\337P .[\251\362|p\371?F\340\37\3708\211[\360(\24C\217\372\341\200\274"..., 2448) = 2448 poll([{fd=4, events=POLLIN|POLLOUT}], 1, 2000) = 1 ([{fd=4, revents=POLLIN}]) write(4, "\27\3\3\0F\250\304\334\237W\235\257\307\276\256fiy\374[\3057\"V\25\332\370V\366<Q\321"..., 75) = -1 EAGAIN (Resource temporarily unavailable)
read(4, "\27\3\3\20\21", 5) = 5read(4, "l z\34\22\221):X7\227\323\5\3702_\246:p\322\302w$\300m\22\242\251\35\213W\360"..., 4113) = 4113 poll([{fd=4, events=POLLIN|POLLOUT}], 1, 2000) = 1 ([{fd=4, revents=POLLIN}]) write(4, "\27\3\3\0F\250\304\334\237W\235\257\307\276\256fiy\374[\3057\"V\25\332\370V\366<Q\321"..., 75) = -1 EAGAIN (Resource temporarily unavailable)
read(4, "\27\3\3\6\222", 5) = 5read(4, "A\207`\364\26\213US\225\360\340\231\31y]\300V\257S\7\202\311\304\25\\\324\315FSc\206\350"..., 1682) = 1682 poll([{fd=4, events=POLLIN|POLLOUT}], 1, 2000) = 1 ([{fd=4, revents=POLLIN}]) write(4, "\27\3\3\0F\250\304\334\237W\235\257\307\276\256fiy\374[\3057\"V\25\332\370V\366<Q\321"..., 75) = -1 EAGAIN (Resource temporarily unavailable)
read(4, "\27\3\3\7\226", 5) = 5read(4, "\332\331\236\353\fx\1\331\261\314T\t\332\203\0\337/\3460\207\307w!\320\3430H\20\232t\304\276"..., 1942) = 1942 poll([{fd=4, events=POLLIN|POLLOUT}], 1, 2000) = 1 ([{fd=4, revents=POLLIN}]) write(4, "\27\3\3\0F\250\304\334\237W\235\257\307\276\256fiy\374[\3057\"V\25\332\370V\366<Q\321"..., 75) = -1 EAGAIN (Resource temporarily unavailable)
read(4, "\27\3\3\20\21", 5) = 5read(4, "\25\353*\200\241\234\0332y\274S#|\325\307\254;\4\244\3772\343i9\232K\213\6K\332?O"..., 4113) = 4113 poll([{fd=4, events=POLLIN|POLLOUT}], 1, 2000) = 1 ([{fd=4, revents=POLLIN}]) write(4, "\27\3\3\0F\250\304\334\237W\235\257\307\276\256fiy\374[\3057\"V\25\332\370V\366<Q\321"..., 75) = -1 EAGAIN (Resource temporarily unavailable)
read(4, "\27\3\3\10\214", 5) = 5read(4, "\206\251\203\10R2{\231B%#\274\342~\22\34\342\241Y\25\206\206Y\227B\272\ncA\277\3033"..., 2188) = 2188 poll([{fd=4, events=POLLIN|POLLOUT}], 1, 2000) = 1 ([{fd=4, revents=POLLIN}]) write(4, "\27\3\3\0F\250\304\334\237W\235\257\307\276\256fiy\374[\3057\"V\25\332\370V\366<Q\321"..., 75) = -1 EAGAIN (Resource temporarily unavailable)
read(4, "\27\3\3\20\21", 5) = 5read(4, "\340\360\366\345\34\222B]l\355V\215\26\317a\7\357E\2143i\356\"\273\335G2\0\263\353K\225"..., 4113) = 4113 poll([{fd=4, events=POLLIN|POLLOUT}], 1, 2000) = 1 ([{fd=4, revents=POLLIN}]) write(4, "\27\3\3\0F\250\304\334\237W\235\257\307\276\256fiy\374[\3057\"V\25\332\370V\366<Q\321"..., 75) = -1 EAGAIN (Resource temporarily unavailable)
read(4, "\27\3\3\20\21", 5) = 5read(4, "n\366\203\331\365\211\203)\266\3272\223\3^\212\3}^\371Z\256\3731_\334C\343\7`\233-\22"..., 4113) = 4113 poll([{fd=4, events=POLLIN|POLLOUT}], 1, 2000) = 1 ([{fd=4, revents=POLLIN}]) write(4, "\27\3\3\0F\250\304\334\237W\235\257\307\276\256fiy\374[\3057\"V\25\332\370V\366<Q\321"..., 75) = -1 EAGAIN (Resource temporarily unavailable)
read(4, "\27\3\3\20\21", 5) = 5read(4, "[\364\317\22\205\276\321\247\215~\236\372\266~B>\21\350\215u\302\347\31\315\356Q8Y\305l\262\237"..., 4113) = 4113 poll([{fd=4, events=POLLIN|POLLOUT}], 1, 2000) = 1 ([{fd=4, revents=POLLIN}]) write(4, "\27\3\3\0F\250\304\334\237W\235\257\307\276\256fiy\374[\3057\"V\25\332\370V\366<Q\321"..., 75) = -1 EAGAIN (Resource temporarily unavailable)
read(4, "\27\3\3\20\21", 5) = 5read(4, "\232#Y\27\351\243\363e=\17\267\356\365CCxar\373\230\322\331\24\315X\322zj\206O\262\257"..., 4113) = 4113 poll([{fd=4, events=POLLIN|POLLOUT}], 1, 2000) = 1 ([{fd=4, revents=POLLIN}]) write(4, "\27\3\3\0F\250\304\334\237W\235\257\307\276\256fiy\374[\3057\"V\25\332\370V\366<Q\321"..., 75) = -1 EAGAIN (Resource temporarily unavailable)
read(4, "\27\3\3\2A", 5) = 5read(4, "\236\345\357\333}\37\374\33\347\355\252(\261P\36\30\210\226f\221\347\360\265xj\215\361\222s6BJ"..., 577) = 577 poll([{fd=4, events=POLLIN|POLLOUT}], 1, 2000) = 1 ([{fd=4, revents=POLLIN}]) write(4, "\27\3\3\0F\250\304\334\237W\235\257\307\276\256fiy\374[\3057\"V\25\332\370V\366<Q\321"..., 75) = -1 EAGAIN (Resource temporarily unavailable)
read(4, "\27\3\3\0\365", 5) = 5read(4, "e+\231s\344,M\tN\344\10\360l\276\365\33AS?\366o\310o\310>l\361l\214P\f\25"..., 245) = 245
lseek(3, 0, SEEK_SET) = 0 write(3, "1591221302\n4727930\n", 19) = 19 ftruncate(3, 19) = 0 fdatasync(3) = 0openat(AT_FDCWD, "/home/uwe/Maildir-work/.lists.linux-arm-kernel/tmp/1631042587.213028_341581.taurus,U=4727930:2,", O_WRONLY|O_CREAT|O_EXCL, 0600) = 8
write(8, "Return-path: <linux-arm-kernel-b"..., 7344) = 7344 fsync(8) = 0 close(8) = 0 utime("/home/uwe/Maildir-work/.lists.linux-arm-kernel/tmp/1631042587.213028_341581.taurus,U=4727930:2,", {actime=1562341670 /* 2019-07-05T17:47:50+0200 */, modtime=1562341670 /* 2019-07-05T17:47:50+0200 */}) = 0 rename("/home/uwe/Maildir-work/.lists.linux-arm-kernel/tmp/1631042587.213028_341581.taurus,U=4727930:2,", "/home/uwe/Maildir-work/.lists.linux-arm-kernel/new/1631042587.213028_341581.taurus,U=4727930:2,") = 0 write(7, "> 4728202 0 4727930\n", 20) = 20 lseek(3, 0, SEEK_SET) = 0 write(3, "1591221302\n4727931\n", 19) = 19 ftruncate(3, 19) = 0 fdatasync(3) = 0openat(AT_FDCWD, "/home/uwe/Maildir-work/.lists.linux-arm-kernel/tmp/1631042587.213028_341582.taurus,U=4727931:2,", O_WRONLY|O_CREAT|O_EXCL, 0600) = 8
write(8, "Return-path: <linux-arm-kernel-b"..., 7517) = 7517 fsync(8) = 0 close(8) = 0 utime("/home/uwe/Maildir-work/.lists.linux-arm-kernel/tmp/1631042587.213028_341582.taurus,U=4727931:2,", {actime=1562337704 /* 2019-07-05T16:41:44+0200 */, modtime=1562337704 /* 2019-07-05T16:41:44+0200 */}) = 0 rename("/home/uwe/Maildir-work/.lists.linux-arm-kernel/tmp/1631042587.213028_341582.taurus,U=4727931:2,", "/home/uwe/Maildir-work/.lists.linux-arm-kernel/new/1631042587.213028_341582.taurus,U=4727931:2,") = 0 write(7, "> 4728203 0 4727931\n", 20) = 20 lseek(3, 0, SEEK_SET) = 0 write(3, "1591221302\n4727932\n", 19) = 19 ftruncate(3, 19) = 0 fdatasync(3) = 0openat(AT_FDCWD, "/home/uwe/Maildir-work/.lists.linux-arm-kernel/tmp/1631042587.213028_341583.taurus,U=4727932:2,", O_WRONLY|O_CREAT|O_EXCL, 0600) = 8
write(8, "Return-path: <linux-arm-kernel-b"..., 7470) = 7470 ... The process has the following mapping of fd numbers to files: lr-x------ 1 uwe uwe 64 Sep 7 21:27 0 -> /dev/null lrwx------ 1 uwe uwe 64 Sep 7 21:27 1 -> 'socket:[669143]' lrwx------ 1 uwe uwe 64 Sep 7 21:27 2 -> 'socket:[669143]'lrwx------ 1 uwe uwe 64 Sep 7 21:27 3 -> /home/uwe/Maildir-work/.lists.linux-arm-kernel/.uidvalidity
lrwx------ 1 uwe uwe 64 Sep 7 21:27 4 -> 'socket:[669937]'l-wx------ 1 uwe uwe 64 Sep 7 21:27 5 -> /home/uwe/Maildir-work/.lists.linux-arm-kernel/.mbsyncstate.lock l-wx------ 1 uwe uwe 64 Sep 7 21:27 6 -> /home/uwe/Maildir-work/.lists.linux-arm-kernel/.mbsyncstate.new l-wx------ 1 uwe uwe 64 Sep 7 21:27 7 -> /home/uwe/Maildir-work/.lists.linux-arm-kernel/.mbsyncstate.journal l-wx------ 1 uwe uwe 64 Sep 7 21:27 8 -> '/home/uwe/Maildir-work/.lists.linux-arm-kernel/tmp/1631042850.213028_347408.taurus,U=4733757:2,'
Maybe it's the server's fault if it scrambled all UIDs?
I saw mbsync write to one of its files at 100% yesterday while the ssh tunnel was already gone.
Currently the ssh tunnel is up.
This was before I suspended the machine, and some time after wakeup today, the loss of 460840 messages was reported from that process.
I added -V to the invokation of mbsync and the log currently has:Sep 07 16:55:41 taurus syncmail-work[212854]: Opening master box lists/linux-arm-kernel... Sep 07 16:55:41 taurus syncmail-work[212854]: Opening slave box lists/linux-arm-kernel...
Sep 07 16:55:42 taurus syncmail-work[212854]: Loading master... Sep 07 16:55:42 taurus syncmail-work[212854]: Loading slave...Sep 07 16:55:45 taurus syncmail-work[212854]: slave: 4386047 messages, 513211 recent Sep 07 16:55:52 taurus syncmail-work[212854]: master: 523665 messages, 2 recent
Sep 07 16:55:52 taurus syncmail-work[212854]: Synchronizing... [...]Sep 07 17:02:41 taurus syncmail-work[213028]: Opening master box lists/linux-arm-kernel... Sep 07 17:02:41 taurus syncmail-work[213028]: Opening slave box lists/linux-arm-kernel...
Sep 07 17:02:42 taurus syncmail-work[213028]: Loading master... Sep 07 17:02:42 taurus syncmail-work[213028]: Loading slave...Sep 07 17:02:44 taurus syncmail-work[213028]: slave: 4386049 messages, 513213 recent Sep 07 17:03:02 taurus syncmail-work[213028]: master: 523670 messages, 523670 recent
Sep 07 17:03:02 taurus syncmail-work[213028]: Synchronizing...the first part is from the previous run which completed successfully a bit later. Now it's 21:38 local time here and mbsync still works on
lists/linux-arm-kernel/ .
ok, now i want to see the -D logs from an event like that, bracketed by timestamps.
I was lazy and didn't save -D logs.
Would it make sense to put my maildir into a git repo (including the mbsync files) and commit after each run?that's an interesting idea. in particular, this could be used to verify that notmuch doesn't mess up anything in the synced folders.
Also this I didn't implement yet ...
Is there a more suitable place where I can read about how mbsync tracks mail than the source code to understand the content of these files?nope. everything is in sync.c. load_state() is probably most helpful, as it also provides insight into the journal's meaning.
ok. It seems I have to do some homework now ... Best regards Uwe
OpenPGP_signature
Description: OpenPGP digital signature
_______________________________________________ isync-devel mailing list isync-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/isync-devel