Closed Bug 81400 Opened 23 years ago Closed 23 years ago

More printfs during startup

Categories

(SeaMonkey :: Build Config, enhancement)

Product:
SeaMonkey
Component:
Build Config
Type:
enhancement
Priority:
Not set
Severity:
normal

Tracking

(Not tracked)

Status:
VERIFIED WONTFIX

People

(Reporter: BenB, Assigned: cls)

Details

It happens from time to time that a Mozilla build just exits (crashes?) for me
during startup. (E.g. right now after "profileName passed in: <name>".) I use
non-debug, optimized builds.

There are some quite long periods (e.g. after the above msg) during startup
where no output at the console is made. If there were more output, it would help
tracking the problem down.

I suggest to place some status msgs in strategical places during startup.

I don't know, which these places are, so -> Browser-General.
Severity: normal → enhancement
ben, they are actually going in the opposite direction (against my wishes) and
removing _all_ messages to the console.  You can watch them go, one at a time by
cc'ing yourself on bug 76720.  
Should this be WONTFIX, then?
we could give this to xpapps.  Personally I agree that we need more debug 
output in some cases because it's really hard to triage general startup 
failures, the best i can do is compare what libraries a good moz loads to what 
libraries a bad moz loads.

Perhaps we should discuss this in a newsgroup?
Assignee: asa → pchen
Component: Browser-General → XP Apps
OS: Linux → All
QA Contact: doronr → sairuh
Hardware: PC → All
->build config...? as the antithesis to bug 76720. however, punt back if it
should really be in xp apps (it might fall through the cracks, however).
Assignee: pchen → cls
Component: XP Apps → Build Config
QA Contact: sairuh → granrose
Do I even have to say it?
Status: NEW → RESOLVED
Closed: 23 years ago
Resolution: --- → WONTFIX
Sorry? While bug 76720 is related, recent discussion there shows that it doesn't
preclude this bug.

What are the reasons why this bug should not be fixed?

If engineers need to fix a crash on startup problem, they can use their debug
bits.  You can reopen this bug if you want (I certainly can't stop you), but I'd
reassign it to someone else because I doubt you will find any Netscape people
interested in working on, or delivering any bits with additional printfs to the
console.
granrose,
the problem is that I, with my tester hat on, don't run debug binaries (for many
reasons - actually, I don't even use DEBUG for my dev builds). If a build exits
without comment during startup, I can't even file a meaningful bug report,
because I have no chance to narrow the problem to any component/reason. Just
filing a report and hoping that anybody with a debug build can reproduce it is
no solution either, because these bug are usually not reproducable by most
others (otherwise they would block the tree opening).

It is fine that you cannot assign any engineers to work on that. But there is a
large difference between "have no time/interest" (-> <nobody@mozilla.org>) and
"should not be fixed" (WONTFIX).
If you mean the latter, please state why. If you mean the former, please reopen
and reassign.

verified.
verified.
Status: RESOLVED → VERIFIED
Product: Browser → Seamonkey
You need to log in before you can comment on or make changes to this bug.