dax
(Daniela)
October 20, 2017, 11:27pm
1
Lately I seem to have some problems with sidekiq jobs.
As on sidekiq I can not see the logs of failed processes it is always a little mystery to understand what did not work.
For example, looking on the dashboard and set the stats for 6 months I can see this graphic:
The chart tends to rise steadily and after about 20000 processes it seems to reset and start over again.
So I was looking for github if there was some update-plugin-something to implement the sidekiq interface and I found this GitHub - davydovanton/sidekiq-statistic: See statistic about your workers .
Has anyone already tested it?
(damned keyboard shortcuts, I saved the post before finishing writing )
4 Likes
dax
(Daniela)
October 21, 2017, 4:33pm
2
We were able to install the gem 'sidekiq statistics' on our test server, I thought there would be more problems. Currently sidekiq is active and working and the tab statistics is visible.
1 Like
tgxworld
(Alan Tan)
October 24, 2017, 7:44am
3
We used to ship the gem with Discourse but it was removed due to
opened 11:38PM - 15 Sep 15 UTC
closed 02:46PM - 01 Feb 21 UTC
bug
https://github.com/davydovanton/sidekiq-statistic/blob/master/lib/sidekiq/statis… tic/middleware.rb#L40
The timeslist should be expired after some amount of time and (I'd suggest) the times aggregated into a much more compact form. The Web UI could store hourly aggregates per worker.
If I were implementing this, I'd compact the timeslists hourly. The compact form could store:
```
execution count
average time
standard deviation
98th percentile
etc
```
So instead of a list of thousands of numbers, it would be converted into a hash with 4-6 numbers per hour. You could expire the hashes after one month, giving the user a nice recent history to compare.
which results in a memory leak in Redis.
6 Likes