Bug 14138 - No log files written to SD card - Jive reboots after suspend if has SD card with logs dir
: No log files written to SD card - Jive reboots after suspend if has SD card w...
Status: CLOSED FIXED
Product: SB Controller
Classification: Unclassified
Component: Diagnostics
: unspecified
: PC Windows XP
: P1 normal with 1 vote (vote)
: 7.5.0
Assigned To: Richard Titmuss
:
Depends on:
Blocks:
  Show dependency treegraph
 
Reported: 2009-09-18 17:41 UTC by Mickey Gee
Modified: 2010-04-08 17:24 UTC (History)
7 users (show)

See Also:
Category: ---


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Mickey Gee 2009-09-18 17:41:47 UTC
Using 7.4 firmware r7649, I was unable to create log files. I don't remember the directory name to create on the SD card, but I tried "log", "logs", and "Log" and none of them worked.

Yes, I verified that the SD card was NOT in read-only mode. The SD card worked fine for taking Controller screen shots.
Comment 1 James Richardson 2009-09-19 10:43:21 UTC
Mickey: please format the card, then put directory LOG on the card, then EJECT the card from the system.

If the cards file system is left 'open' then you can't right to it.  This normally happens if you 'pull' the card from the system with out ejecting it.
Comment 2 Marc Auslander 2009-09-20 16:44:16 UTC
I have the same problem - log not written to the card.  I don't understand the comment from James.

First a fact.  I can ssh to the controller and write on the card.

I don't know what the eject business is about.

Are you saying the card needs to be reformatted?  Why - it works fine as is.

I suspect the issue is a race between running rcS and udev getting the card mounted. This has become an issue in later version of linux.  The test for the card in rcS "works" once the controller is up.
Comment 3 James Richardson 2009-09-21 08:11:45 UTC
Marc: in the past, I was told this information.  It looks like something new is happening however.  Thanks for the extra debugging.

Richard: care to investigate this one, or assign it to another engineer and target please.
Comment 4 Mickey Gee 2009-09-21 14:29:19 UTC
I did as suggested in Comment 1. Still no logs.
Comment 5 James Richardson 2009-09-23 09:26:27 UTC
Richard: can you have a look at this post 7.4
Comment 6 Marc Auslander 2009-09-25 05:41:10 UTC
I tried to verify my race hypothesis.  A sleep of 10 in rcS just before the check for a log directory on the card causes the log to go on the card!  sleep 4 does it sometime!

BUT - with the log on the card, jive won't run.  It keep rebooting.  There is stuff written on the log.  I won't copy it here since you can get it when you get around to looking at this.

Obviously not a high priority.
Comment 7 James Richardson 2009-10-07 10:37:13 UTC
*** Bug 14487 has been marked as a duplicate of this bug. ***
Comment 8 Marc Auslander 2009-10-10 08:21:39 UTC
This bug is more important than I realized.  My controller (7.4.1 latest) just rebooted spontaneously.  Since the internal log is on ramdisk and gets restarted each boot, there is no way to see what caused the reboot!  So I urge more urgency for fixing the log on card problems.
Comment 9 Wadzinski Tom 2009-10-15 08:25:16 UTC
From James regarding the reboot issue:
SD card with Log folder, wait for Jive goes to suspend, pick up jive to wake = reboot
SD card without Log folder, wait for Jive goes to suspend, pick up jive to wake = wakes normally
No SD card, wait for Jive goes to suspend, pick up jive to wake = wakes normally
Comment 10 Mickey Gee 2009-10-15 19:25:18 UTC
Are there two issues in this bug?

1. Won't write to logs directory on SD card
2. Reboots after suspend if SD card with logs directory in root

If so, there should be 2 bugs and not one.
Comment 11 Richard Titmuss 2009-11-02 11:44:32 UTC
== Auto-comment from SVN commit #7991 to the jive repo by richard ==
 == https://svn.slimdevices.com/jive?view=revision&revision=7991 ==

Fixed Bug #14523
While booting wait for mmc/sd cards or usb drives to be mounted before
continuing. This allows the debug logging to work correctly.

Mount removable volumes without sync, otherwise it's too slow.
Comment 12 Chris Owens 2010-04-08 17:24:54 UTC
This bug has been marked fixed in a released version of Squeezebox Server or the accompanying firmware or mysqueezebox.com release.

If you are still seeing this issue, please let us know!