Add confirmation checkbox to enable array Stop button


Recommended Posts

not a bug, but something i ran into today.

 

when a parcheck/correct is running can you please disable the array stop button. the wording is such that I see it and immediately think "oh, thats the stop parcheck button". but then realize after i press it that "cancel" is the proper button.

 

just a minor annoyance. ;)

 

LimeTech edit: The original Subject was "Disable array STOP button if parity check is active."

Link to comment

not a bug, but something i ran into today.

 

when a parcheck/correct is running can you please disable the array stop button. the wording is such that I see it and immediately think "oh, thats the stop parcheck button". but then realize after i press it that "cancel" is the proper button.

 

just a minor annoyance. ;)

 

+1000

 

Sent from my Galaxy Nexus using Tapatalk 2

 

 

Link to comment

not a bug, but something i ran into today.

 

when a parcheck/correct is running can you please disable the array stop button. the wording is such that I see it and immediately think "oh, thats the stop parcheck button". but then realize after i press it that "cancel" is the proper button.

 

just a minor annoyance. ;)

 

I can't tell you how many times I have done this!!!

 

John

Link to comment

not a bug, but something i ran into today.

 

when a parcheck/correct is running can you please disable the array stop button. the wording is such that I see it and immediately think "oh, thats the stop parcheck button". but then realize after i press it that "cancel" is the proper button.

 

just a minor annoyance. ;)

 

I can't tell you how many times I have done this!!!

 

John

 

+1, These are the call it small things that should be ironed out by final

Link to comment

not a bug, but something i ran into today.

 

when a parcheck/correct is running can you please disable the array stop button. the wording is such that I see it and immediately think "oh, thats the stop parcheck button". but then realize after i press it that "cancel" is the proper button.

 

just a minor annoyance. ;)

 

I can't tell you how many times I have done this!!!

 

John

 

These are the call it small things that should be ironed out by final, can the posts be moved to the issues list for ease of monitoring progress for this valid request/bug/incompleteness (whatever one chooses to call it)

 

this drives me nuts when i do it. the system becomes unresponsive and I have to telnet in and issue stop commands for the add-ons currently active. If I dont do that fast enough the whole thing becomes lifeless.

 

:'(

 

either disable the button during parcheck or add a checkbox that you need like a reboot/shutdown. either will accomplish the goal of preventing accidentally pushing this button during a parcheck when one is trying to stop the parcheck.

 

Link to comment
  • 1 month later...

Pause parity button too?

 

Sent from my GT-I9505 using Tapatalk 4

Why?

 

Yeah.. may seem like a stupid thing, but I was trying to copy some TV show's to a USB hard drive for my buddy during a parity check and it is painfully slow during a check obviously.

 

I ended up stopping the check, but it was 75% complete. It would have been nice to be able to pause it.

Link to comment
  • 3 weeks later...

as said before, the stop button disable/ checkbox is a minor annoyance, but i hit it also before... so that's a +1

as far as the pause parity check goes - i am all for it!!!! i wished to have THAT option several times already! sure, with a little planning one can work around it....  ::)

so, i am for it, IF and that's a huge IF... it is easy to implement at the going builds. otherwise i would opt for consideration with the next version, not taking time away from progress!!!

 

my 2c

Lars

Link to comment

Personally I think a pause parity would be great, especially considering many of us have very large drives that take a LONG time to complete a parity check. While I have a check scheduled for night time when I'm least likely to be using the server, especially to a degree that the transfer rates would be an issue, things do happen and I have had to cancel parity checks that were still hours away from being finished but well over the halfway point.

Link to comment
  • 2 weeks later...

I also like the idea -- but it's not as simple as just saving the state and then continuing => any activity that resulted in writes to that part that had already been checked would require re-checking that section of the parity disk.

 

Of course this is also true for writes during an active parity check -- but the logic would likely be somewhat different if the check was inactive (e.g. paused).

 

The reality is that it's not that big a deal to simply Stop a check when you need to use the array for unexpected activities, and then just start it back up when it's more convenient.    This isn't, after all, something that's done frequently enough to be a major hassle.

 

Link to comment

Parts that had already been checked but has been written to during the pause wouldn't necessarily need to be rechecked. Unless I'm mistaken the parity check doesn't currently do that.

 

For example, if I'm running a parity check that is 90% complete and content is added to the beginning of the disk that has already been checked, the parity check doesn't go back and recheck the changed section.

 

If it doesn't do it currently while running, no reason it must if its been paused.

Link to comment

That's true, if that's the way it worked.    I thought that an active parity check would re-check any parity writes that happened during the check -- but if that's not true, then a Pause that behaved the same way would be fine.

 

What that means, however, is that a parity check that finishes with zero errors doesn't really mean there are no errors if there has been any write activity on the array during the check (since there's no read verification done on writes).    A very minor point -- but nevertheless something to consider if you're doing a parity check to confirm a clean array before doing some activity that results in "running at risk".  (e.g. replacing a drive with a larger one)

 

Link to comment

well presumably you wrote the correct parity during that write unless something really odd happened.

 

Clearly we always hope that parity was written correctly -- but it is NOT verified by UnRAID after the update.    My point is simply that if you're going to do something that depends on parity being good, it's a good idea to be sure you have a true zero sync error parity disk => and that means doing a parity check without doing ANY writes during the check.

 

Note that the same thing is true if you do a check and it finds and corrects sync errors => that's why you should always run a 2nd check immediately afterwards, to confirm the corrections were all successful.

 

Link to comment

I don't see the pause option as necessary as a checkbox for the stop button. Like the power down option has.

 

I think pausing the parity check could cause problems atleast with the unraid newbies. I too did not see the full importance of that function (parity check) in the beginning. I have never seen those failed writes on a pc, if there have been any. But this thing is different.

 

The stop button is so close to the refresh button that there has been some "heart-stopping" moments when my tired eyes have been trying to focus correctly. Is it really that much cpu-heavy to get it update like every 10 min or something like that?

 

Sometimes it gets extremely slow and I like to see the estimate in real-time. But atleast if I try to refresh I would like not to accidentially stop the process.

 

Edit: IIRC simple features can do real-time refresh on parity check...

 

 

Link to comment

If this gets implemented, think it could be useful also if able to resume same position even after shutdown/re-start ... useful for systems that take really long time to do parity check... on situations that server needs to be turned off for some reason...

Link to comment