Opened 5 years ago

Last modified 5 years ago

#10946 new bug

[package_daemon] crashed while running haikuporter

Reported by: diver Owned by: bonefish
Priority: normal Milestone: R1
Component: Servers/package_daemon Version: R1/Development
Keywords: Cc:
Blocked By: Blocking:
Has a Patch: no Platform: All

Description

hrev47361 gcc2h.

Had a mounted chrooted volume (left from building this package before) which haikuporter couldn't unmount: unmount: unmounting failed: Device/File/Resource busy

I force unmounted it with: unmount -f /data/haikuports/sys-libs/libexecinfo/work-1.1/boot/system

Then I ran haikuporter libexecinfo and a second later package_daemon crashed.

Attachments (1)

package_daemon-197-debug-14-06-2014-23-01-44.report (13.5 KB ) - added by diver 5 years ago.

Download all attachments as: .zip

Change History (3)

comment:1 by diver, 5 years ago

syslog snippet before the crash:

KERN: bfs: bfs_create_symlink:1017: File or Directory already exists
KERN: Last message repeated 5 times.
KERN: packagefs [93176026552: 13146] Adding packages from "/data/haikuports/sys-libs/libexecinfo/work-1.1/boot/system/packages"
KERN: packagefs: Failed to get shine-through directory "non-packaged": No such file or directory
KERN: packagefs: Failed to get shine-through directory "var": No such file or directory
KERN: packagefs [93176029694: 13146] KERN: StringPool usage:
KERN: packagefs [93176031297: 13146]   total unique strings:       29894,   386904 bytes, overhead: 448410 bytes
KERN: packagefs [93176031765: 13146]   total strings with dups:   153123,  1622292 bytes
KERN: packagefs [93176032156: 13146]   unshared strings:            1533
KERN: packagefs [93176032446: 13146]   bytes saved:               786978
KERN: package_daemon [93176033384:   197] root at "/data/haikuports/sys-libs/libexecinfo/work-1.1/boot" (device: 8, node: 3146714) registered
KERN: package_daemon [93176033974:   197] volume at "/data/haikuports/sys-libs/libexecinfo/work-1.1/boot/system" registered
KERN: 13147: DEBUGGER: getNumAvailable() == 0
KERN: debug_server: Thread 13147 entered the debugger: Debugger call: `getNumAvailable() == 0'
KERN: stack trace, current PC 0x62096114  commpage_syscall + 0x4:
KERN:   (0x726e1b08)  0x1295a25  __assert_fail + 0x51
KERN:   (0x726e1b38)  0x1307f28  malloc__Q28BPrivate10threadHeapUl + 0x48c
KERN:   (0x726e1b88)  0x1308864  malloc + 0x184
KERN:   (0x726e1bb8)  0x1a27ceb  _Allocate__7BStringl + 0x27
KERN:   (0x726e1be8)  0x1a27e21  _Clone__7BStringPCcl + 0x21
KERN:   (0x726e1c18)  0x1a27ddc  _Init__7BStringPCcl + 0x28
KERN:   (0x726e1c48)  0x1a22d9a  __7BStringPCc + 0x4a

in reply to:  description comment:2 by bonefish, 5 years ago

Replying to diver:

Had a mounted chrooted volume (left from building this package before) which haikuporter couldn't unmount: unmount: unmounting failed: Device/File/Resource busy

A while back I added printing more diagnosis info (the open FDs) for this case. It would be nice, if you could create a ticket with the info attached when you encounter it again.

Regarding this issue, it apparently is a heap corruption. Given that there's at least one other ticket (#10898) that suggests some heap corruption issue, I hope sifting through the Coverity issues will turn something up.

Note: See TracTickets for help on using tickets.