dashy
dashy copied to clipboard
[BUG] Using 66% of a CPU when idle
Environment
Self-Hosted (Docker)
System
Ubuntu 22.0.4
Version
2.1.1
Describe the problem
With no browser windows taking to dashy, the "node server" process is using 66% of a CPU core at minimum. Presumably this is background processing such as the docker healthcheck. There's no rebuild running. It's not going to zero at any point. With any browser window open, it goes up to 125% and stays there or above. However, this all seems very excessive for what's logically happening.
This was a straight-forward docker install using Portainer on a Raspberry Pi 4 running Ubuntu 22.04. This is enough CPU use to trigger the cooling fan on this box, hence why I've noticed it.
Additional info
partial output of pstree -a -p:
ββcontainerd-shim,654085 -namespace moby -id fcfa6e2acda69da0fbd4ca458e75243f3b7b4563f64322e231e6566505c71980 -address /run/containerd/containerd.sock
β ββtini,654108 -- yarn build-and-start
β β ββnode,654156 /opt/yarn-v1.22.15/bin/yarn.js build-and-start
β β ββnode,654179 /app/node_modules/.bin/npm-run-all --parallel build-watch start
β β β ββnode,654190 /opt/yarn-v1.22.15/bin/yarn.js run build-watch
β β β β ββnode,654212 /app/node_modules/.bin/vue-cli-service build --watch --mode production
β β β β β ββ{node},654214
β β β β β ββ{node},654215
β β β β β ββ{node},654216
β β β β β ββ{node},654217
β β β β β ββ{node},654218
β β β β β ββ{node},654224
β β β β β ββ{node},654232
β β β β β ββ{node},654233
β β β β β ββ{node},654234
β β β β β ββ{node},654235
β β β β ββ{node},654192
β β β β ββ{node},654193
β β β β ββ{node},654194
β β β β ββ{node},654195
β β β β ββ{node},654196
β β β β ββ{node},654202
β β β β ββ{node},654204
β β β β ββ{node},654205
β β β β ββ{node},654206
β β β β ββ{node},654207
β β β ββnode,654191 /opt/yarn-v1.22.15/bin/yarn.js run start
β β β β ββnode,654213 server
β β β β β ββ{node},654219
β β β β β ββ{node},654220
β β β β β ββ{node},654221
β β β β β ββ{node},654222
β β β β β ββ{node},654223
β β β β β ββ{node},654225
β β β β β ββ{node},654226
β β β β β ββ{node},654227
β β β β β ββ{node},654228
β β β β β ββ{node},654229
β β β β ββ{node},654197
β β β β ββ{node},654198
β β β β ββ{node},654199
β β β β ββ{node},654200
β β β β ββ{node},654201
β β β β ββ{node},654203
β β β β ββ{node},654208
β β β β ββ{node},654209
β β β β ββ{node},654210
β β β β ββ{node},654211
β β β ββ{node},654180
β β β ββ{node},654181
β β β ββ{node},654182
β β β ββ{node},654183
β β β ββ{node},654184
β β β ββ{node},654185
β β β ββ{node},654186
β β β ββ{node},654187
β β β ββ{node},654188
β β β ββ{node},654189
β β ββ{node},654167
β β ββ{node},654168
β β ββ{node},654169
β β ββ{node},654170
β β ββ{node},654171
β β ββ{node},654172
β β ββ{node},654175
β β ββ{node},654176
β β ββ{node},654177
β β ββ{node},654178
β ββ{containerd-shim},654086
β ββ{containerd-shim},654087
β ββ{containerd-shim},654088
β ββ{containerd-shim},654089
β ββ{containerd-shim},654090
β ββ{containerd-shim},654091
β ββ{containerd-shim},654092
β ββ{containerd-shim},654093
β ββ{containerd-shim},654094
β ββ{containerd-shim},654375
β ββ{containerd-shim},657704
Please tick the boxes
- [X] You have explained the issue clearly, and included all relevant info
- [X] You are using a supported version of Dashy
- [X] You've checked that this issue hasn't already been raised
- [X] You've checked the docs and troubleshooting guide
- [X] You agree to the code of conduct
If you're enjoying Dashy, consider dropping us a β
π€ I'm a bot, and this message was automated
Same problem here... I'm about to look for an alternative because my raspberry freezes and I have to restart it.
I had the same problem, so I built a new docker image from master, and that sort of seems to have fixed it. However, I also noticed that when I had the status check enabled, it would interfere with my pi-alert container (Note that pi-alert is VERY buggy). And both would consume a lot of cpu.
This issue has gone 6 weeks without an update. To keep the ticket open, please indicate that it is still relevant in a comment below. Otherwise it will be closed in 5 working days.
AFAIK, nothing has changed. Sounds like Lissy93 is swamped.
This issue has gone 6 weeks without an update. To keep the ticket open, please indicate that it is still relevant in a comment below. Otherwise it will be closed in 5 working days.
It is still relevant, no stale
This issue has gone 6 weeks without an update. To keep the ticket open, please indicate that it is still relevant in a comment below. Otherwise it will be closed in 5 working days.
Still relevant. @Lissy93, any chance you'll be able to look at this? If not, perhaps you have some hints as to how you'd diagnose it?
Hey, Really sorry for the delay in replying, (I have so many unread GH notifications at the moment, that it's hard to spot the important ones! π¬)
This definitely doesn't sound good at all, weirdly I'm running a very similar setup (Pi 4, default Docker, etc) and no issues. Looking at those logs, I'm wondering if it's something gone wrong with the healthcheck. If you disable that, do you still see issues?
@ohmantics, do you have an example config file to reproduce this issue? Also, does it still happen if you change the container command to run yarn start
instead?
@Lissy93, is there any reason for the ARM-based Docker images to have yarn build-and-start
as default CMD, while the Docker image for x86 is just yarn start
?
This issue has gone 6 weeks without an update. To keep the ticket open, please indicate that it is still relevant in a comment below. Otherwise it will be closed in 5 working days.
No stale
This issue has gone 6 weeks without an update. To keep the ticket open, please indicate that it is still relevant in a comment below. Otherwise it will be closed in 5 working days.
It is still relevant, no stale
Same issue with docker compose, Dashy v2.1.1 on Debian 11 in a raspberry pi4. I was searching what was using 100% of CPU constantly until I stopped the stack. It does't go down when Dashy isn't accessed.
Same issue with docker compose, Dashy v2.1.1 on Debian 11 in a raspberry pi4. I was searching what was using 100% of CPU constantly until I stopped the stack. It does't go down when Dashy isn't accessed.
Turning off Enable status check in page settings and restarting the container fixed it. I think I'm going to use uptime kuma for now
This issue has gone 3 months without an update. To keep the ticket open, please indicate that it is still relevant in a comment below. Otherwise it will be closed in 5 working days.
Still relevant.
This issue has gone 3 months without an update. To keep the ticket open, please indicate that it is still relevant in a comment below. Otherwise it will be closed in 5 working days.
Still relevant
I also run Dashy with many other things on my Rpi 3B+ with 1GB RAM and Dashy is consuming very few MB and barely any CPU even when starting up.
Can you confirm if you are running x64 or x32 ?
This issue was automatically closed because it has been stalled for over 1 year with no activity.
Auto-closing bots are a project management smell.
Auto-closing bots are a project management smell.
I haven't gotten any confirmation that it is still relevant, so the bot did everything correct.
Still relevant
Then please answer the question I asked above.
Instead of ranting on the bot!
In my case x64
@bdovaz @ohmantics
Hi Could you test this with the :latest tag by repulling and rebuilding the app?
Keep in mind V3 has breaking changes: https://github.com/Lissy93/dashy/releases/tag/3.0.0
Not seeing the original issue now. It did take several re-pulls for the build to complete without errors; not sure why since it says "possible error, check logs" and there are no logs to see.