Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Issues with "Reboot Later" #1174

Closed
1 of 2 tasks
azulskyknight opened this issue Jun 15, 2022 · 11 comments
Closed
1 of 2 tasks

Issues with "Reboot Later" #1174

azulskyknight opened this issue Jun 15, 2022 · 11 comments
Assignees
Labels
bug Something isn't working

Comments

@azulskyknight
Copy link
Contributor

Server Info (please complete the following information):

  • OS: [Ubuntu 20.04 LTS and Debian 11.3]
  • Browser: [Firefox 101.0.1 64bit, Chrome 102.5005.115 64bit, New Edge 102.0.1245.39 64bit ]
  • RMM Version (as shown in top left of web UI): v0.13.4

Installation Method:

  • Standard
  • Docker

Agent Info (please complete the following information):

  • Agent version (as shown in the 'Summary' tab of the agent from web UI): v2.0.4
  • Agent OS: Windows 10 v21H1, Windows 10 v21H2, Windows Server 2016 and 2019

Describe the bug
Firefox displays a different UI tool to select the date time stamp passed for the scheduled reboot. Firefox only shows the calendar portion, no time selector is available.

If a reboot is scheduled for a time that has already passed, a 400 timeout error is returned. This may be intentional, but seems like it should be handled a little more gracefully.

During the scheduling process, if any of the elements are typed in manually the scheduling event returns a 400 invalid date error.

If on Firefox, once this corruption is revealed the only way to clear it is to refresh the browser and try again without actually typing anything in. Which limits the user to changing the date alone.

If on a Chromium browser the mangled date time can be fixed by clicking the day on the calendar intended. If a manual date is typed in, both today and that date will be selected on the calendar, and clicking the correct day will result in a working time stamp that successfully schedules.

On both browsers if the seconds value on the time is reduced to 00, it will simply vanish once a date is chosen on the calendar. Firefox doesn't display the time selector tools at all and will return an invalid date error when such a shortened timestamp is used. Chromium based browsers lose the seconds column in the time selector portion of the UI but seem to schedule properly assuming a date is selected on the calendar as a final action first.

To Reproduce
Steps to reproduce the behavior:

  1. Click an agent
  2. right click, reboot -> later
  3. Type values into the date and time fields. (without using the UI)
  4. Click Schedule Reboot

(See above for using the UI to "fix" things)

Expected behavior
I expect to be able to type in a proper date / time string, or select one from the UI and have it work the first time. It would also be nice if I could copy / paste the entire date time string into the reboot time box so I can rapidly schedule multiple reboots quickly.

Screenshots
image
image
image
image

Additional context
Silversword was assisting me (AzulSkyknight) in #Support on Discord, searching for myself for comments from today 6-15-2022 around 9am Arizona time will find the conversation.

@azulskyknight
Copy link
Contributor Author

Please note, this behavior existed on two different tacticalrmm installations, the first was Ubuntu 20.04 LTS based, and the second Debian 11. I rebuilt my server onto the Debian OS over the weekend in an effort to clear up some lingering redis issues. Redis is now quite happier! But the UI bug remains. :)

@silversword411 silversword411 added the bug Something isn't working label Jun 16, 2022
@wh1te909 wh1te909 added this to To do in Ticket Triage via automation Jun 22, 2022
@eCloudJTuttle
Copy link

Using chrome developer tools to inspect the patch command, the date/time being passed is invalid format.
image

wh1te909 added a commit that referenced this issue Jul 18, 2022
@wh1te909
Copy link
Member

I have fixed the 400 timeout error when date is in past in e8e19fe (this was actually causing the agent to crash LOL which is why the timeout).

As for the invalid date issue, I am having a hard time reproducing it, I have tried typing the values by hand instead of using the GUI but don't get any errors. Only time I get error is if I type an invalid year. Any way to record a GIF or video of how to consistently reproduce? Thanks.

@silversword411
Copy link
Collaborator

Those are screenshots from firefox...different date picker than chrome, but I can't get it to do it either now. Maybe it's fixed?

@wh1te909
Copy link
Member

Yes I tried both firefox and chrome, probably was fixed in a quasar update

@eCloudJTuttle
Copy link

I can also confirm that the issue has gone away in chrome.

@silversword411
Copy link
Collaborator

Post if found to be a problem still

Ticket Triage automation moved this from To do to Done Jul 18, 2022
@azulskyknight
Copy link
Contributor Author

azulskyknight commented Aug 2, 2022

Issue still exists, at least in small part. (v0.14.5)

Please note the behavior is drastically improved, it's far more consistent now and typing into the date time field doesn't break the UI nearly as frequently.

There are circumstances I cannot fully duplicate with New Edge and Firefox where the time field doesn't contain seconds. The UI is passing a timestamp missing those digits, instead of as zeros. As a result, invalid timestamp and 400 error.

image

You can see here Firefox is still using a different picker, and because it lacks the time it cannot add the seconds to the time required to "work around" this issue.

New Edge simply doesn't give you the seconds selector when the time has no seconds too, which also results in invalid timestamps being passed.

Firefox date only selector:
image

New Edge Can work showing selector:
image

Again New Edge will at times vanish the seconds counter, but not every time. I was able to schedule a reboot at 0 seconds just a moment ago. Might be old browser cache garbage?

@silversword411 silversword411 reopened this Aug 2, 2022
Ticket Triage automation moved this from Done to In progress Aug 2, 2022
Ticket Triage automation moved this from In progress to Done Aug 5, 2022
@wh1te909
Copy link
Member

wh1te909 commented Aug 9, 2022

Fixed (again lol) in Release 0.14.6

@JSuenram
Copy link

Same issue here in v0.15.7, did not test in v0.15.9...

@silversword411
Copy link
Collaborator

silversword411 commented Apr 25, 2023

Maybe reboot has the same TZ logic problem as 9f95c57 ?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working
Projects
Development

No branches or pull requests

6 participants