| 1 |
Jason Stubbs wrote: |
| 2 |
> On Thursday 01 December 2005 10:27, Zac Medico wrote: |
| 3 |
> |
| 4 |
>>Brian Harring wrote: |
| 5 |
>> |
| 6 |
>>>>2) Display a warning message via an atexit hook when parallel-fetching is |
| 7 |
>>>>enabled, in order to alert the user that background fetching may _still_ |
| 8 |
>>>>be in progress if emerge appears to hang after an ebuild dies (this |
| 9 |
>>>>happened to me while kde-3.5 was fetching in the background). |
| 10 |
>>> |
| 11 |
>>>Details please... |
| 12 |
>>>atexit portage_exec hooks should kill off the fetcher. |
| 13 |
>> |
| 14 |
>>Well, I'm running the latest svn so there could be a regression in the |
| 15 |
>>recent changes to portage_exec.cleanup(). |
| 16 |
> |
| 17 |
> |
| 18 |
> Does changing SIKTERM to SIGKILL fix it? |
| 19 |
> |
| 20 |
|
| 21 |
No, SIGKILL doesn't seem to make a difference (not a regression then). It seems that the atexit hooks do not execute until the sys.exit() is called in the fetcher child process (see attached test case that reproduces the problem). Perhaps it would be more appropriate to implement parallel-fetch with a thread rather than a fork? |
| 22 |
|
| 23 |
Zac |
Archives