Memory Leak in forge-pull? #698
Replies: 2 comments 6 replies
-
I'm afraid that is a wontfix. Forge 0.4 requires Emacs 27.1. It might kinda work with older releases, but the use of an ancient release could very well be the cause of this issue. One result of this is that the new builtin and module back-ends of Emacsql cannot be used. Instead the legacy custom external process back-end has to be used. That back-end will be removed soon, possibly as early as Monday. Additionally support for Gitlab is limited. Much fewer Forge users use it with Gitlab than with Github, so I cannot justify putting as much effort into supporting the former. I recommend you update to Emacs 29.4. That might just by-pass this issue altogether (and has many other benefits). |
Beta Was this translation helpful? Give feedback.
-
Off-topic: @t-zuehlsdorff My emails to you no longer go through for some reason. |
Beta Was this translation helpful? Give feedback.
-
Aloha,
thank you for your release! I was finally able to use forge successfully in several small projects and it is great! :)
However, i found a bug when trying to use a mid-size project. Doing a
forge-pull
in small projects with ~ 100 issues and MRs is fine. But when i do this for ~ 1.200 issues and MRs emacs crashes. I can see that the issues got fetched and the MRs are getting pulled in. At some point in time my Laptop become slower, later emacs will not render anymore and get killed.I saw it eating up to 32 GB of RAM before the OOM-Killer did his job.
I am using forge 0.4.1 on FreeBSD and emacs 24.2. I am pulling from a self hosted GitLab.
I am not sure how i can provide helpful information to debug this issue. If you can guide me i will provide everything you need :)
Best,
Torsten
Beta Was this translation helpful? Give feedback.
All reactions