Thunderbird fails to start with blocklist.xml: TypeError: doc is null
Categories
(Toolkit :: Blocklist Implementation, defect)
Tracking
()
| Tracking | Status | |
|---|---|---|
| firefox-esr68 | 76+ | fixed |
| firefox74 | --- | unaffected |
| firefox75 | --- | unaffected |
| firefox76 | --- | unaffected |
| firefox77 | --- | unaffected |
People
(Reporter: mozilla-bugs, Assigned: Gijs)
Details
Attachments
(1 file)
|
47 bytes,
text/x-phabricator-request
|
RyanVM
:
approval-mozilla-esr68+
|
Details | Review |
User Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) QtWebEngine/5.14.1 Chrome/77.0.3865.129 Safari/537.36
Steps to reproduce:
I had Thunderbird crashing, and then it failed to start.
On the terminal I've seen:
JavaScript error: resource://gre/modules/Blocklist.jsm, line 2086: TypeError: doc is null
I've found that there is an empty blocklist.xml file (0 bytes).
Removing it allowed Thunderbird to start again.
This should be handled more gracefully.
Updated•6 years ago
|
Updated•6 years ago
|
Comment 1•6 years ago
|
||
Although this bug was seen with Thunderbird, I'm reassigning it to "toolkit". Looks like this is in general code not specific to TB.
| Assignee | ||
Comment 2•6 years ago
|
||
Ugh. The XML implementation has been removed from mozilla-central. It's not on any supported branches anymore besides ESR 68. :-\
I suppose a nullcheck isn't particularly hard to add, though it does seem like quite the edgecase.
| Assignee | ||
Comment 3•6 years ago
|
||
Updated•6 years ago
|
| Assignee | ||
Comment 4•6 years ago
|
||
Comment on attachment 9138591 [details]
Bug 1618375 - handle empty blocklist.xml file in case of crashes, r?florian
ESR Uplift Approval Request
- If this is not a sec:{high,crit} bug, please state case for ESR consideration: Adds a nullcheck and prevents Firefox/Thunderbird never starting up
- User impact if declined: Some users may get stuck in a state where Firefox/Thunderbird never starts up
- Fix Landed on Version: n/a
- Risk to taking this patch: Low
- Why is the change risky/not risky? (and alternatives if risky): I'm just adding some nullchecks. We're handling a case where previously, the
loadevent handler would throw an exception, so we'd wait forever on aPromise, and potentially not start up as a result. - String or UUID changes made by this patch: n/a
| Assignee | ||
Comment 5•6 years ago
|
||
Ryan, this code has been removed on central/beta (and is preffed off on release so landing it there makes no difference) so it cannot really land anywhere other than esr. Please can you ensure this shows up on the relevant person's triage for approval? Thanks!
| Assignee | ||
Comment 6•6 years ago
|
||
(In reply to :Gijs (he/him) from comment #5)
Ryan, this code has been removed on central/beta (and is preffed off on release so landing it there makes no difference) so it cannot really land anywhere other than esr. Please can you ensure this shows up on the relevant person's triage for approval? Thanks!
Ryan says I don't need to do anything else. :-)
Comment 7•6 years ago
|
||
Comment on attachment 9138591 [details]
Bug 1618375 - handle empty blocklist.xml file in case of crashes, r?florian
Add a null check to avoid crashing in some edge cases. Approved for 68.8esr.
Updated•6 years ago
|
Comment 8•6 years ago
|
||
| bugherder uplift | ||
Updated•6 years ago
|
Comment 9•6 years ago
|
||
Thanks for uplifting this.
Just for grins I tried this on WIndows to see if I could get a crash report: 68.6.0 starts up, 68.4.1 just hangs forever with a little CPU
Description
•