Bug 118480 - [umass] Timeout in USB mass storage freezes vfs layer or crashes
Summary: [umass] Timeout in USB mass storage freezes vfs layer or crashes
Status: Open
Alias: None
Product: Base System
Classification: Unclassified
Component: usb (show other bugs)
Version: 7.0-BETA4
Hardware: Any Any
: Normal Affects Only Me
Assignee: freebsd-usb (Nobody)
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2007-12-07 08:50 UTC by Oliver Adler
Modified: 2017-12-31 22:36 UTC (History)
1 user (show)

See Also:


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Oliver Adler 2007-12-07 08:50:00 UTC
	During backup of two external machines via command(s)
	$ sudo tar cf - . | ssh giga gzip \> /mnt/oli/backup/guppy.tgz
	$ sudo tar cf - . | ssh giga gzip \> /mnt/oli/backup/buster.tgz
	
	After some hours the following sequence occurred in the oplog:
	Dec  7 00:28:52 giga kernel: umass1: BBB reset failed, TIMEOUT
	Dec  7 00:48:22 giga kernel: umass1: BBB bulk-in clear stall failed, TIMEOUT
	Dec  7 00:48:22 giga kernel: umass1: BBB bulk-out clear stall failed, TIMEOUT
	Dec  7 00:48:22 giga kernel: umass1: BBB reset failed, TIMEOUT
	Dec  7 00:48:22 giga kernel: umass1: BBB bulk-in clear stall failed, TIMEOUT
	Dec  7 00:48:22 giga kernel: umass1: BBB bulk-out clear stall failed, TIMEOUT
	Dec  7 00:48:22 giga kernel: umass1: BBB reset failed, TIMEOUT
	Dec  7 00:48:22 giga kernel: umass1: BBB bulk-in clear stall failed, TIMEOUT
	Dec  7 00:48:22 giga kernel: umass1: BBB bulk-out clear stall failed, TIMEOUT
	Dec  7 00:48:22 giga kernel: umass1: BBB reset failed, TIMEOUT
	Dec  7 00:48:22 giga kernel: umass1: BBB bulk-in clear stall failed, TIMEOUT
	Dec  7 00:48:22 giga kernel: umass1: BBB bulk-out clear stall failed, TIMEOUT
	Dec  7 00:48:22 giga kernel: umass1: BBB reset failed, TIMEOUT
	Dec  7 00:48:22 giga kernel: umass1: BBB bulk-in clear stall failed, TIMEOUT
	Dec  7 00:48:22 giga kernel: umass1: BBB bulk-out clear stall failed, TIMEOUT
	Dec  7 00:48:22 giga kernel: g_vfs_done():da4s1d[WRITE(offset=65536, length=2048)]error = 5
	Dec  7 00:48:22 giga kernel: g_vfs_done():da4s1d[WRITE(offset=6144000, length=16384)]error = 5
	Dec  7 00:48:22 giga kernel: g_vfs_done():da4s1d[WRITE(offset=6160384, length=16384)]error = 5
	Dec  7 00:48:22 giga kernel: g_vfs_done():da4s1d[WRITE(offset=6176768, length=2048)]error = 5
	Dec  7 00:48:22 giga kernel: g_vfs_done():da4s1d[WRITE(offset=200978087936, length=16384)]error = 5
	Dec  7 00:48:22 giga kernel: g_vfs_done():da4s1d[WRITE(offset=201018703872, length=131072)]error = 5
	Dec  7 00:48:22 giga kernel: g_vfs_done():da4s1d[WRITE(offset=201170780160, length=16384)]error = 5
	Dec  7 00:48:22 giga kernel: g_vfs_done():da4s1d[WRITE(offset=201363472384, length=16384)]error = 5
	Dec  7 00:48:22 giga kernel: g_vfs_done():da4s1d[WRITE(offset=201556164608, length=16384)]error = 5
	Dec  7 00:48:22 giga kernel: g_vfs_done():da4s1d[WRITE(offset=201748856832, length=16384)]error = 5
	Dec  7 00:48:22 giga kernel: g_vfs_done():da4s1d[WRITE(offset=201941549056, length=16384)]error = 5
	Dec  7 00:48:22 giga kernel: g_vfs_done():da4s1d[WRITE(offset=202134241280, length=16384)]error = 5
	Dec  7 00:48:22 giga kernel: g_vfs_done():da4s1d[WRITE(offset=202326933504, length=16384)]error = 5
	Dec  7 00:48:22 giga kernel: g_vfs_done():da4s1d[WRITE(offset=202334044160, length=131072)]error = 5
	Dec  7 00:48:22 giga kernel: g_vfs_done():da4s1d[WRITE(offset=202334175232, length=131072)]error = 5
	Dec  7 00:48:22 giga kernel: g_vfs_done():da4s1d[WRITE(offset=202334306304, length=131072)]error = 5
	Dec  7 00:48:22 giga kernel: g_vfs_done():da4s1d[WRITE(offset=202334437376, length=131072)]error = 5
	Dec  7 00:48:22 giga kernel: g_vfs_done():da4s1d[WRITE(offset=202334568448, length=131072)]error = 5
	Dec  7 00:48:22 giga kernel: g_vfs_done():da4s1d[WRITE(offset=202334699520, length=131072)]error = 5
	Dec  7 00:48:22 giga kernel: g_vfs_done():da4s1d[WRITE(offset=202334830592, length=131072)]error = 5
	Dec  7 00:48:22 giga kernel: g_vfs_done():da4s1d[WRITE(offset=202334961664, length=131072)]error = 5
	Dec  7 00:48:22 giga kernel: g_vfs_done():da4s1d[WRITE(offset=330467278848, length=16384)]error = 5
	Dec  7 00:48:22 giga kernel: g_vfs_done():da4s1d[WRITE(offset=202333913088, length=131072)]error = 5
	Dec  7 01:11:50 giga syslogd: kernel boot file is /boot/kernel/kernel
	Dec  7 01:11:50 giga kernel: Copyright (c) 1992-2007 The FreeBSD Project.
	After this, the computer could still be pinged, but every
	shell or ssh access was not functioning any more. I suppose
	the vfs layer was dead. To get it working again, I had to
	switch off.

	I tried the same setting once again and got after around one
	hour:
	The machine crached with:
		    Panic String: ffs_clusteralloc: map mismatch

	If you need more info (e.g. vmcore from 2nd run), please give
	me a note.

How-To-Repeat: 	Using the following load it crashed after around one hour:
	Two concurrent tar from external machines, on the target
	machine a trafshow, two gzip (from external tar via ssh):

	$ sudo tar cf - . | ssh giga gzip \> /mnt/oli/backup/guppy.tgz
	$ sudo tar cf - . | ssh giga gzip \> /mnt/oli/backup/buster.tgz

	If you need some more tests, I would like to support by testing
	with a patched kernel or potential different hardware.
Comment 1 Eitan Adler freebsd_committer freebsd_triage 2017-12-31 07:59:51 UTC
For bugs matching the following criteria:

Status: In Progress Changed: (is less than) 2014-06-01

Reset to default assignee and clear in-progress tags.

Mail being skipped