Remove SleepTimer as it is error prone and rely on Pusher instead

[marcaaron]

If you haven’t already, check out our contributing guidelines for onboarding and email contributors@expensify.com to request to join our Slack channel!

---

Action Performed:

Normal app usage

I've been online without any changes in network status, and noticed E.cash on desktop web had the "syncing" icon. I looked and saw countless pages of this -- what is it doing?

Expected Result:

The app should only re-fetch data when necessary

Actual Result:

@quinthar noticed that the desktop app was "syncing" for no reason and saw this in the logs:

• These logs mean that the desktop/web app is repeatedly "polling for data" for some reason.

Workaround:

Yes, but this could create a particularly bad situation if this were to happen on mobile for a user with a large number of reports as we'd be firing off many unnecessary API requests.

Platform:

Web
Desktop App

Version Number:
Logs: https://stackoverflow.com/c/expensify/questions/4856
Notes/Photos/Videos: Any additional supporting documentation
Expensify/Expensify Issue URL:

View all open jobs on Upwork

[MelvinBot]

Triggered auto assignment to @stephanieelliott (AutoAssignerTriage), see https://stackoverflow.com/c/expensify/questions/4749 for more details.

[marcaaron]

I think we're still looking for a reproduction on this one.

[marcaaron]

I noticed this on dev and turned verbose logging on. Pretty sure there is something wrong with the clock timer logic we have based on this...

It exists to make sure we refetch reports when re-opening a laptop, but maybe there's a better way to go about it (or something has recently changed about that code - since I don't recall having this problem until now).

[stephanieelliott]

@marcaaron are we able to repro this yet? I've tried doing it myself (unsuccessful) but not too sure what steps to follow.

[marcaaron]

Nope, I've seen it pop up while working in E.cash locally - but still unsure what causes it to get into that state.

[stephanieelliott]

OK, I'm gonna hold off on passing this through to Engineering until we can repro this.

Or rather than trying to repro ourselves, another idea could be to post reproducing the problem as a job on Upwork?

[marcaaron]

We still don't really have a precedent for that as far as I know, but it's something @mallenexpensify has floated before. It would be nice to get someone working on this.

I did manage to find this Slack thread.

And this older issue that we closed that seems like the same issue.

[mallenexpensify]

I wonder, if for issues like this, where there is an issue reproducing, if we can state at the top of the post need to be able to reproduce' and outline steps before proposing solution, then we just start it like any other issue, and double the price until someone first figures out how to reproduce and proposes a solution. Whatcha think?

[stephanieelliott]

I like that idea @mallenexpensify! I think it could be worth at least trying. Do you think just changing the title and giving it a shout in #expensify-open-source would be the best way to start?

[quinthar]

FYI this is 100% reproducible for me: just leave E.cash open in a tab while you are doing something else, without touching it or doing anything at all, with the console open. After a few minutes the log will be full of these lines:

[quinthar]

Again:

[marcaaron]

So I left my chat open for a bit to see if this was happening and it did, but not with the same frequency and not in an infinite loop or anything.

Looked at the network request logs and noticed a few weird things...

1. There are some calls to Get&returnValueList=reportStuff with an undefined value for reportIDList (doesn't seem like there is any reason to call this)
2. We are logging to the console when there is no IOU report for a chat, but this isn't really an error.

Probably we don't need to log this. But also, why are we calling GetIOUReport for every single report?
Since adding Get&returnValueList=reportSummaryList we should be able to tell if a report has an IOU report or not before calling this which seems like a good idea.

Anyways, I don't think that has anything to do with this issue.

I checked my logs and saw lots of these

66efcd6e7a9af4c4-SJC | www4.rno | 2021-07-15 03:11:16 296 | marc@expensifail.com | [info] [NetworkConnection] Firing reconnection callbacks because timer clock skewed ~~ parameter: '[]' userAgent: 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.114 Safari/537.36'
66efcf0d2983f4c4-LAX | www3.lax | 2021-07-15 03:12:22 668 | marc@expensifail.com | [info] [NetworkConnection] Firing reconnection callbacks because timer clock skewed ~~ parameter: '[]' userAgent: 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.114 Safari/537.36'
66efd06c795bf4c4-LAX | www4.lax | 2021-07-15 03:13:18 866 | marc@expensifail.com | [info] [NetworkConnection] Firing reconnection callbacks because timer clock skewed ~~ parameter: '[]' userAgent: 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.114 Safari/537.36'

So yes, seems like sleep timer stuff is not working correctly and the code for this is here.