[PR #1879] [MERGED] refactor: open boltdb only while using it #2168

New issue

Closed

opened 2026-03-01 21:54:18 +03:00 by kerem · 0 comments

kerem commented

2026-03-01 21:54:18 +03:00

Owner

📋 Pull Request Information

Original PR: https://github.com/nektos/act/pull/1879
Author: @ChristopherHX
Created: 6/25/2023
Status: ✅ Merged
Merged: 7/10/2023
Merged by: @mergify[bot]

Base: master ← Head: boltdb-lock-in-handler

📝 Commits (10+)

a87f85f refactor: open boltdb only while using it
2c42fff patch
2edba96 Update handler_test.go
dc86141 Update handler_test.go
b874083 Update handler_test.go
afe0e8e Update handler.go
2119a66 timeout * 10
c937f87 Merge branch 'master' into boltdb-lock-in-handler
5911237 pr feedback
8b995f8 fixup

📊 Changes

2 files changed (+83 additions, -42 deletions)

View changed files

📝 pkg/artifactcache/handler.go (+79 -40)
📝 pkg/artifactcache/handler_test.go (+4 -2)

📄 Description

I don't know anything about boltdb, but running two act instances with the same bolt database returns after 5s with error timeout.

My proposal is open the db on demand. Additionally close the db during download, upload and commit to reduce hitting a 5s timeout due to large caches.

Do you know any other solution?
This might have a negative impact for a single act instance, due to additional open and close calls.

Does boltdb have any problem when calling Close() twice? If so we need to disable the defered Close() after calling it.

~~Timeout is now 50s to allow using cache in parallel on slower devices, since database locks might take longer.~~

Closes #1825

_{🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.}

## 📋 Pull Request Information **Original PR:** https://github.com/nektos/act/pull/1879 **Author:** [@ChristopherHX](https://github.com/ChristopherHX) **Created:** 6/25/2023 **Status:** ✅ Merged **Merged:** 7/10/2023 **Merged by:** [@mergify[bot]](https://github.com/apps/mergify) **Base:** `master` ← **Head:** `boltdb-lock-in-handler` --- ### 📝 Commits (10+) - [`a87f85f`](https://github.com/nektos/act/commit/a87f85fbbee38143beff5b0d2337c3fb0c2a4e0d) refactor: open boltdb only while using it - [`2c42fff`](https://github.com/nektos/act/commit/2c42fff956bf995f84824ec8e9dd9182a2d50dd1) patch - [`2edba96`](https://github.com/nektos/act/commit/2edba96276771fa3c2d4e9469df2cb30b55b5c28) Update handler_test.go - [`dc86141`](https://github.com/nektos/act/commit/dc86141f720a9536ca1a6f043e5c44b0d65e2eff) Update handler_test.go - [`b874083`](https://github.com/nektos/act/commit/b87408363592abd20f4f57e5b6c4315ac40275f0) Update handler_test.go - [`afe0e8e`](https://github.com/nektos/act/commit/afe0e8ef78800371fedccaba94c33c017c096c43) Update handler.go - [`2119a66`](https://github.com/nektos/act/commit/2119a66655c17f9f379a5030bad4f0be5a41f387) timeout * 10 - [`c937f87`](https://github.com/nektos/act/commit/c937f87be6eebc0540ac1abe1ae86a38364e775c) Merge branch 'master' into boltdb-lock-in-handler - [`5911237`](https://github.com/nektos/act/commit/5911237ce9617821b92c7bf21f1689acfa16b36f) pr feedback - [`8b995f8`](https://github.com/nektos/act/commit/8b995f816a83c2ee5c203e7ee2a34d5088e735ae) fixup ### 📊 Changes **2 files changed** (+83 additions, -42 deletions) <details> <summary>View changed files</summary> 📝 `pkg/artifactcache/handler.go` (+79 -40) 📝 `pkg/artifactcache/handler_test.go` (+4 -2) </details> ### 📄 Description I don't know anything about boltdb, but running two act instances with the same bolt database returns after 5s with error timeout. My proposal is open the db on demand. Additionally close the db during download, upload and commit to reduce hitting a 5s timeout due to large caches. Do you know any other solution? This might have a negative impact for a single act instance, due to additional open and close calls. Does boltdb have any problem when calling Close() twice? If so we need to disable the defered Close() after calling it. ~~Timeout is now 50s to allow using cache in parallel on slower devices, since database locks might take longer.~~ Closes #1825 --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>

kerem

2026-03-01 21:54:18 +03:00

closed this issue
added the
pull-request
label

kerem referenced this issue

2026-03-01 21:54:51 +03:00

[PR #2168] [MERGED] Add containerd's normalized architectures to archMapper #2299

No milestone

No project

No assignees

1 participant

Notifications

Due date

The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference

starred/act#2168

No description provided.

Rows
Columns