Opened 9 years ago

Closed 9 years ago

Last modified 9 years ago

#2063 closed defect (fixed)

Vidalia has no clue why Tor crashed, and says that it crashed because of ""

Reported by: nickm Owned by:
Priority: Medium Milestone:
Component: Archived/Vidalia Version:
Severity: Keywords:
Cc: chiiph@… Actual Points:
Parent ID: Points:
Reviewer: Sponsor:

Description

See bug #2062. Tor is crashing (or dying, or asserting, or just not there), and Vidalia is saying,

Please report the following error message to the Tor developers at bugs.torproect.org: ""

So there are at least two problems here. One is that Vidalia doesn't know what happened, and isn't doing a good job of finding out. The other problem is that instead of saying "I couldn't tell why", it acts as if the empty string is an error message. Not good.

Child Tickets

Change History (10)

comment:1 Changed 9 years ago by chiiph

It seems this could either be because of bad parsing the response message from tor, or its a tor problem not giving a reason for the error. I would say it's a tor related problem, since the parsing function is used for every type of event, and this seems to be the only problem with them.

A minor fix for this is to check whether the reason is empty, and replace that with "There's been an unknown problem with Tor, please fill a bug report and attach the contents of <path_to_the_advanced_log>".

Opinions?

comment:2 Changed 9 years ago by nickm

This message is generated from what kind of event? If there's something we should give more info on from Tor, let's do that.

comment:3 Changed 9 years ago by chiiph

This message is generated by this status event (from control-spec):
BUG

"REASON=STRING"

Tor has encountered a situation that its developers never expected,
and the developers would like to learn that it happened. Perhaps
the controller can explain this to the user and encourage her to
file a bug report?

{Controllers should log bugs, but shouldn't annoy the user in case a
bug appears frequently.}

And REASON seems to be empty in this case.

comment:4 Changed 9 years ago by nickm

Well that's odd. Looking at the code, the BUG status event is generated in control.c in control_event_logmsg. The reason is taken straight from the log message that we were about to log in logv in log.c. I added some asserts to make sure it's really there, and I can't figure out why it wouldn't be.

comment:5 Changed 9 years ago by chiiph

Cc: chiiph@… added

nickm: Were you able to reproduce this with the new asserts?

comment:6 Changed 9 years ago by nickm

I've not heard of anyone trying; it would be good to know if they turn up, though.

comment:7 Changed 9 years ago by nickm

I have reports that this could be related to #1125.

comment:8 Changed 9 years ago by chiiph

Resolution: fixed
Status: newclosed

I'm considering this one fixed since the related Tor-side ticket is.

comment:9 Changed 9 years ago by rransom

#2062 was closed as a duplicate of #1998, which is still open.

comment:10 Changed 9 years ago by chiiph

I was talking about #1125.

Note: See TracTickets for help on using tickets.