about:memory displays a weird error due to lacking access to /proc/self/smaps

Bug #911540 reported by Micah Gersten
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Mozilla Firefox
Invalid
Low
firefox (Ubuntu)
Triaged
Low
Unassigned

Bug Description

Please note, this only affects users when the apparmor profile is enabled.
"An error occurred when collecting a multi-reporter's results: [Exception... \"Component returned failure code: 0x80004005 (NS_ERROR_FAILURE) [nsIMemoryMultiReporter.collectReports]\" nsresult: \"0x80004005 (NS_ERROR_FAILURE)\" location: \"JS frame :: chrome://global/content/aboutMemory.js :: getReportersByProcess :: line 295\" data: no]"

The error should go away on Ubuntu after bug 911494 is fixed. Bug 911494 is about apparmor blocking access to /proc/self/smaps. Upstream seems interested in refining this message.

This bug is about refining the message rather than making it go away.

Revision history for this message
In , Mozilla-bugs-micahscomputing (mozilla-bugs-micahscomputing) wrote :

I currently get the following error at the top of about:memory on Firefox 9.0.1 and Gecko/20111229 Firefox/12.0a1 (Both Ubuntu builds):
"An error occurred when collecting a multi-reporter's results: [Exception... \"Component returned failure code: 0x80004005 (NS_ERROR_FAILURE) [nsIMemoryMultiReporter.collectReports]\" nsresult: \"0x80004005 (NS_ERROR_FAILURE)\" location: \"JS frame :: chrome://global/content/aboutMemory.js :: getReportersByProcess :: line 295\" data: no]"

https://bugzilla.mozilla.org/show_bug.cgi?id=688233#c23 suggested this was desired, so I filed a bug

Revision history for this message
In , Mozilla-bugs-micahscomputing (mozilla-bugs-micahscomputing) wrote :

Sorry, I forgot to put in the first comment that I suggest telling the user something like 'Unable to access /proc/self/smaps' per that other bug comment.

Changed in firefox:
importance: Unknown → Medium
status: Unknown → Confirmed
Changed in firefox:
importance: Medium → Low
Revision history for this message
In , Nicholas Nethercote (n-nethercote) wrote :

The smaps reading code is long gone, so this shouldn't be a problem any more.

Changed in firefox:
status: Confirmed → Invalid
Revision history for this message
Daniel Richard G. (skunk) wrote :

For the record, Firefox 61 very much does continue to make use of /proc/NNNN/smaps (and /proc/NNNN/statm) when using the about:memory page.

I confirmed this by experimentally commenting out the AppArmor rules for those two /proc files (motivated by comment #3 above), and subsequently observed AppArmor denials. However, the about:memory page still displayed the memory summary, and no error message came up at all.

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.