Bug 54623 - journalctl don't follow logs
Summary: journalctl don't follow logs
Status: RESOLVED WORKSFORME
Alias: None
Product: systemd
Classification: Unclassified
Component: general (show other bugs)
Version: unspecified
Hardware: Other All
: medium normal
Assignee: systemd-bugs
QA Contact: systemd-bugs
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2012-09-07 07:05 UTC by Oleksii Shevchuk
Modified: 2012-11-01 17:47 UTC (History)
0 users

See Also:
i915 platform:
i915 features:


Attachments

Description Oleksii Shevchuk 2012-09-07 07:05:46 UTC
> journalctl -n 1  -fa
Logs begin at Thu, 23 Aug 2012 05:34:28 +0300.
Sep 07 10:06:36 CLU kernel: usb 3-1: usbfs: process 9305 (nfc-eventd) did not claim interface 0 before use
Sep 07 10:07:03 CLU kernel: usb 3-1: reset full-speed USB device number 6 using uhci_hcd
Sep 07 10:07:04 CLU kernel: usb 3-1: reset full-speed USB device number 6 using uhci_hcd
Sep 07 10:07:04 CLU nfc-eventd[9305]: NFC reader: ACS / ACR122U PICC Interface opened
Sep 07 10:07:04 CLU nfc-eventd[9305]: Found MIFARE Classic card:
Sep 07 10:07:04 CLU nfc-eventd[9305]: ATQA (SENS_RES): 00  04
Sep 07 10:07:04 CLU nfc-eventd[9305]: UID (NFCID1): 77  c6  80  43
Sep 07 10:07:04 CLU nfc-eventd[9305]: SAK (SEL_RES): 88
Sep 07 10:07:04 CLU nfc-eventd[9305]: Guessing size: seems to be a 1024-byte card
Sep 07 10:07:06 CLU kernel: usb 3-1: usbfs: process 9305 (nfc-eventd) did not claim interface 0 before use
Sep 07 10:07:07 CLU kernel: usb 3-1: reset full-speed USB device number 6 using uhci_hcd
Sep 07 10:07:08 CLU kernel: usb 3-1: reset full-speed USB device number 6 using uhci_hcd
Sep 07 10:07:10 CLU kernel: usb 3-1: usbfs: process 9305 (nfc-eventd) did not claim interface 0 before use
Sep 07 10:07:11 CLU kernel: usb 3-1: reset full-speed USB device number 6 using uhci_hcd


 > journalctl -n 20  -fa
Logs begin at Thu, 23 Aug 2012 05:34:28 +0300.
Sep 07 10:07:13 CLU nfc-eventd[9305]: Duplicate. Removing last
Sep 07 10:07:16 CLU kernel: usb 3-1: reset full-speed USB device number 6 using uhci_hcd
Sep 07 10:07:16 CLU kernel: usb 3-1: reset full-speed USB device number 6 using uhci_hcd
Sep 07 10:07:17 CLU nfc-eventd[9305]: NFC reader: ACS / ACR122U PICC Interface opened
Sep 07 10:07:17 CLU nfc-eventd[9305]: Found MIFARE Classic card:


and so on

i.e. I pass -f option, jorunalctl follows some records, then do nothing

with strace i get something like that:
read(3, "\2\0\0\0\2\0\0\0\0\0\0\0 \0\0\0user-1000.journa"..., 4112) = 48
read(3, 0xbf8676bc, 4112)               = -1 EAGAIN (Resource temporarily unavailable)
Comment 1 Lennart Poettering 2012-09-12 07:26:13 UTC
Is this one some non-local FS? I.e. NFS or so? Or is this plain ext4?
Comment 2 Lennart Poettering 2012-09-12 07:26:30 UTC
Also: which version of systemd is this?
Comment 3 Oleksii Shevchuk 2012-09-12 08:34:48 UTC
Ext4, local FS, systemd 189
Comment 4 Lennart Poettering 2012-09-13 17:00:58 UTC
What do you mean by "do nothing"? It just waits for more messages. If you generate messages with "logger", do they show up?

What precisely is this bug report about?

What did you expect what didn't happen? What happened that you didn't expect to?
Comment 5 Oleksii Shevchuk 2012-09-13 18:42:27 UTC
logger hangs and don't show new messages. After I restart it, i can see, that there were some new records. I can easily reproduce this:

// log is alias for journalctl

> date
Чт. сент. 13 21:42:34 EEST 2012

> log -fa
Logs begin at Sun, 09 Sep 2012 08:16:11 +0300.
Sep 13 21:26:54 CLU systemd[31328]: display@home.service failed to run 'start-pre' task: No such file or directory
Sep 13 21:26:54 CLU sh[23008]: GET: http://informer.gismeteo.ru/xml/34300_1.xml
Sep 13 21:26:55 CLU sh[23008]: GET: http://informer.gismeteo.ru/xml/33345_1.xml
Sep 13 21:26:55 CLU sh[23008]: GET: http://informer.gismeteo.ru/xml/99894_1.xml
Sep 13 21:31:44 CLU pulseaudio[31354]: [pulseaudio] sink.c: Default and alternate sample rates are the same.
Sep 13 21:42:32 CLU nfc-eventd[23257]: nfc-eventd 0.1.5
Sep 13 21:42:33 CLU kernel: systemd-journal[19492]: segfault at b73fc010 ip b77c49c3 sp bfe19470 error 4 in systemd-journald[b77ac000+2c000]
Sep 13 21:42:33 CLU systemd-coredump[23264]: Detected coredump of the journal daemon itself, diverting coredump to /var/lib/systemd/coredump/.
Sep 13 21:42:40 CLU kernel: usb 3-1: reset full-speed USB device number 10 using uhci_hcd
Sep 13 21:42:40 CLU kernel: usb 3-1: reset full-speed USB device number 10 using uhci_hcd
Sep 13 21:42:43 CLU kernel: usb 3-1: usbfs: process 23257 (nfc-eventd) did not claim interface 0 before use
Sep 13 21:42:44 CLU kernel: usb 3-1: reset full-speed USB device number 10 using uhci_hcd
Sep 13 21:42:45 CLU kernel: usb 3-1: reset full-speed USB device number 10 using uhci_hcd
^C

At this moment no messages arrived, till

> date
Чт. сент. 13 21:42:58 EEST 2012

--> 21:42:58 <---

You can see, that after restarting, I can get some messages, that not arrived on the previous run

--> 21:42:48 <---

> log -fa
Logs begin at Sun, 09 Sep 2012 08:16:11 +0300.
Sep 13 21:42:48 CLU kernel: usb 3-1: reset full-speed USB device number 10 using uhci_hcd
Sep 13 21:42:49 CLU kernel: usb 3-1: reset full-speed USB device number 10 using uhci_hcd
Sep 13 21:42:51 CLU kernel: usb 3-1: usbfs: process 23257 (nfc-eventd) did not claim interface 0 before use
Sep 13 21:42:52 CLU kernel: usb 3-1: reset full-speed USB device number 10 using uhci_hcd
Sep 13 21:42:53 CLU kernel: usb 3-1: reset full-speed USB device number 10 using uhci_hcd
Sep 13 21:42:55 CLU kernel: usb 3-1: usbfs: process 23257 (nfc-eventd) did not claim interface 0 before use
Sep 13 21:42:57 CLU kernel: usb 3-1: reset full-speed USB device number 10 using uhci_hcd
Sep 13 21:42:57 CLU kernel: usb 3-1: reset full-speed USB device number 10 using uhci_hcd
Sep 13 21:43:00 CLU kernel: usb 3-1: usbfs: process 23257 (nfc-eventd) did not claim interface 0 before use
Sep 13 21:43:01 CLU kernel: usb 3-1: reset full-speed USB device number 10 using uhci_hcd
Sep 13 21:43:02 CLU kernel: usb 3-1: reset full-speed USB device number 10 using uhci_hcd
Sep 13 21:43:04 CLU kernel: usb 3-1: usbfs: process 23257 (nfc-eventd) did not claim interface 0 before use
Comment 6 Oleksii Shevchuk 2012-09-25 14:46:07 UTC
Couldn't reproduce with latest git
Comment 7 Oleksii Shevchuk 2012-10-06 17:37:10 UTC
Present at current git. Got it after ~6 hours of following
Comment 8 Oleksii Shevchuk 2012-10-09 08:45:03 UTC
waiting at

gdb$ bt
#0  0xb7744424 in __kernel_vsyscall ()
#1  0xb75844d3 in __GI___poll (fds=fds@entry=0xbf9999b0, nfds=nfds@entry=0x1, timeout=0xffffffff) at ../sysdeps/unix/sysv/linux/poll.c:83
#2  0xb776f9c4 in fd_wait_for_event (fd=0x3, event=event@entry=0x1, t=0xffffffffffffffff) at src/shared/util.c:4461
#3  0xb77772bb in sd_journal_wait (j=0xb8197028, timeout_usec=0xffffffffffffffff) at src/journal/sd-journal.c:1977
#4  0xb776be6d in main (argc=0x2, argv=0xbf999d04) at src/journal/journalctl.c:916

> ls -l /proc/$(pidof journalctl)/fd/3
lr-x------ 1 avatar 64 окт.   9 11:47 /proc/4890/fd/3 -> anon_inode:inotify
Comment 9 Lennart Poettering 2012-10-17 17:19:56 UTC
Fixed in git. (hope that it is fixed for good this time...)


Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.