| Time |
Nickname |
Message |
|
03:23
🔗
|
|
vampurr has quit IRC (Read error: Operation timed out) |
|
03:32
🔗
|
|
vampurr has joined #warrior |
|
05:58
🔗
|
|
bithippo has quit IRC (Textual IRC Client: www.textualapp.com) |
|
09:37
🔗
|
|
nertzy has joined #warrior |
|
10:08
🔗
|
|
nertzy has quit IRC (Quit: This computer has gone to sleep) |
|
19:42
🔗
|
JAA |
Would it make sense to have a default item timeout limit in seesaw which could be overridden on a per-project basis? (Or does that already exist?) |
|
19:43
🔗
|
JAA |
That would avoid eternal loops from blocking pipelines forever. |
|
19:44
🔗
|
JAA |
s/avoid/prevent/ |
|
19:45
🔗
|
JAA |
I think there are several slots blocked due to the UOL forums project currently, for example. I've seen a few items looping forever due to --tries inf and invalid status 200 replies (e.g. truncated body), which doesn't get caught by the Lua script. |
|
20:26
🔗
|
Kaz |
would be ideal, as far as I'm aware it doesn't exist |
|
21:54
🔗
|
JAA |
Turns out that there is something that's used in the warrior (but not when running scripts): https://github.com/ArchiveTeam/seesaw-kit/blob/dd6674908bd5bafae8b6011970912bc6535c39c0/seesaw/warrior.py#L803-L806 |
|
21:54
🔗
|
JAA |
This reboots the warrior after a week, it seems. |
|
21:54
🔗
|
JAA |
Which effectively means that any item can only take one week at most. |
|
21:55
🔗
|
JAA |
Ah no, it tries a graceful shutdown after a week and then a forced shutdown 2 days later. |
|
21:55
🔗
|
JAA |
So 9 days at most. |