Mouse click refusing to "unclick".

I do VNC locally via SSH over the host-only network, so no other hardware required. However, with hires-screens there is a noticeable delay until text is "unblurrred".
In my observation during 2 months he mouse remains stuck for days, not for minutes.

 

OK so wait a minute. Are you saying that you VNC into the VM on the same physical machine? If so, nice trick. Would like to know how to do that. Or are you saying that you have a second machine on the local area network in the same location as you, as opposed to screen sharing into a machine out there on the interwebs somewhere?

 

I do this on a single physical machine. For security and to avoid trouble with DHCP-based Ip addresses I use a separate network for the communication between VMs, i.e. a host-only network which can be managed from the native machine with ipconfig. If you leave out the SSH part (no extra security) then this is all to be done.

 

OK, so I'm learning something here. I assume my local UNIX guy would know what all this means? And you don't experience any mouse lockup because the mouse is, in effect, talking to the host OS at all times, right?

 

yes.

 

I am also having this issue with my Mac VM. If it is not resolved soon I will have to switch back to VMWare Fusion. This is a showstopper issue. One has to be able to reliably left-click.

 

Guys, can you please check if suspending/resuming the VM releases the mouse click? Go to Actions > Suspend. Once the VM is suspended, resume the VM via the same context menu. If not, please go to Help menu > Send Technical Data... > generate a report and post the ID here please.

 

@Dmitry@Parallels. As the original poster, I thank you for your response... finally. But before I say anything else you need to know I'm a very unhappy customer at this point. Let's do a recap...

1. I originally posted about this issue SEVEN MONTHS ago. I also called your phone support multiple times to report it and they kept directing me back here.
2. As you can see from other replies it's not isolated to me.
3. It does seem to be isolated to Mac guests (not Windows) and it seems those of us using Parallels for the purposes of using Mac guests are in the minority. Every time I've spoken to Parallels support they tell me to do something to Windows. I've figured out what they're really trying to say is to do something to the guest VM, but it's infuriating when I've repeatedly pointed out that it's a *****Mac***** OS X guest. It's hard to believe anyone's listening.
4. For example. Hemnath@Parallels on Nov 25 (more than a month after I posted the issue - why so long??) told us to shut down Windows re-install Parallels, then try a couple of other things. That seemed to solve it for one commenter, but a lot of other commenters are still posting since then that that didn't solve it and the issue persists - myself included. But again... Windows?? No... it's a Mac guest!!!
5. Now... 6 more months after that happened (including another detailed post from me in January) and finally someone from Parallels responds.
6. According to this page: https://kb.parallels.com/en/124262, in the time since I posted this, Parallels has been publicly updated three times: 13.2.0-43213, 13.3.0-43321, 13.3.1-43365. I had reported the issue by phone at some point before the 13.1.1-43120 release, and then posted this forum message when I found that that release didn't fix it.

Here's the thing. Sure, if you guys are having a hard time getting to the bottom of the issue, then fine. I'm a software developer. I get bugs, especially intermittent ones, are difficult to track down and fix. But Since Henmath's post in November last year, 6 months have gone past, with 17 more comments to this post, and only now does someone at Parallels pay any attention.

This bug isn't some little inconvenience. This bug makes Parallels Desktop for Mac 13 UNUSABLE. I haven't needed to virtualize a Mac in that time, and so I've been quietly hoping you'd fix it by the time I needed to again. But today I've needed it again, and it's still not fixed. It's a handy coincidence that someone from Parallels (you) finally responded a few days before I needed that, otherwise I'd be calling for a refund today instead of writing this.

Now, you've responded. You've asked two things:

1. Does suspending and resuming the VM release the mouse click?
2. If not please send a technical report and post the ID here.

Ok. The answer to your first question is yes. Suspending and resuming the VM releases the mouse click. The VM is usable again until the next time the issue occurs. In the last hour of using a Mac VM today, the issue has occurred four times. Ok... is that Unusable? Yes and no. I'm trying to use it as a FileMaker Server. If I have to suspend and resume the machine every 15 minutes, that means kicking off my users to do so. That's not an option. So in my case it's unusable. But even without it being a server, having to do that every 15 minutes or less... barely usable. And the competition (VMWare Fusion) has issues of its own (that are what made me switch to Parallels 8 months ago in the first place) but their issues don't make it unusable, so that's looking a lot more attractive at this point. Yes, Parallels has some features VMWare Fusion doesn't, that I'd really like to stick with, but if this issue doesn't get resolved soon, I'll be done and I'll make that call for a refund.

To your second question: given that suspending the VM and resuming it resolves the issue, it's hard to know if a technical report can help you, since you only said to send the technical report "If not...". Needless to say I did create one just now anyway, as the issue recurred, in case it might help. The ID is 258542314. The technical report behind that ID was made while the mouse is stuck in the VM and can't be released. I encountered the issue, made the report, suspended the VM, resumed the VM, and then the VM is working again. I then tried to create another technical report immediately, but on trying to submit it, Parallels Desktop crashed ("unexpectedly quit"). I reopened Parallels Desktop again and tried again. Same thing. I now can't send a technical report without it crashing. I suppose I'll have to reinstall the application again. If I have any success after that, I will post more information here.

I sincerely hope you guys can resolve this issue very very soon. Again, I can understand if the issue takes time to fix, but when you're not even communicating anything, meaning you're not even working on it, that's a deal breaker. For an issue that makes the application unusable, I'd think you'd pay a little more attention. If you'll reply to me soon I will work with you on this. If I don't hear anything back soon enough I will call for a refund.

[To others with the same issue, I'd recommend a similar strategy. Please give Dmytri what he's asked for and let's do everything we can at our end to help get this issue resolved. But if they don't respond promptly to that, then after all of us call for a refund, perhaps then they'll listen.]

 

A week later and nothing? Again?

 

Me too - everything DavidT18 wrote in May and June. PLEASE GET BACK TO US!!!

Quoting DavidT18, since his issue and experience is identical to mine:


... Let's do a recap...

1. I originally posted about this issue SEVEN MONTHS ago. I also called your phone support multiple times to report it and they kept directing me back here.
2. As you can see from other replies it's not isolated to me.
3. It does seem to be isolated to Mac guests (not Windows) and it seems those of us using Parallels for the purposes of using Mac guests are in the minority. Every time I've spoken to Parallels support they tell me to do something to Windows. I've figured out what they're really trying to say is to do something to the guest VM, but it's infuriating when I've repeatedly pointed out that it's a *****Mac***** OS X guest. It's hard to believe anyone's listening.
4. For example. Hemnath@Parallels on Nov 25 (more than a month after I posted the issue - why so long??) told us to shut down Windows re-install Parallels, then try a couple of other things. That seemed to solve it for one commenter, but a lot of other commenters are still posting since then that that didn't solve it and the issue persists - myself included. But again... Windows?? No... it's a Mac guest!!!
5. Now... 6 more months after that happened (including another detailed post from me in January) and finally someone from Parallels responds.
6. According to this page: https://kb.parallels.com/en/124262, in the time since I posted this, Parallels has been publicly updated three times: 13.2.0-43213, 13.3.0-43321, 13.3.1-43365. I had reported the issue by phone at some point before the 13.1.1-43120 release, and then posted this forum message when I found that that release didn't fix it.

Here's the thing. Sure, if you guys are having a hard time getting to the bottom of the issue, then fine. I'm a software developer. I get bugs, especially intermittent ones, are difficult to track down and fix. But Since Henmath's post in November last year, 6 months have gone past, with 17 more comments to this post, and only now does someone at Parallels pay any attention.

This bug isn't some little inconvenience. This bug makes Parallels Desktop for Mac 13 UNUSABLE. I haven't needed to virtualize a Mac in that time, and so I've been quietly hoping you'd fix it by the time I needed to again. But today I've needed it again, and it's still not fixed. It's a handy coincidence that someone from Parallels (you) finally responded a few days before I needed that, otherwise I'd be calling for a refund today instead of writing this.

Now, you've responded. You've asked two things:

1. Does suspending and resuming the VM release the mouse click?
2. If not please send a technical report and post the ID here.

Ok. The answer to your first question is yes. Suspending and resuming the VM releases the mouse click. The VM is usable again until the next time the issue occurs. In the last hour of using a Mac VM today, the issue has occurred four times. Ok... is that Unusable? Yes and no. I'm trying to use it as a FileMaker Server. If I have to suspend and resume the machine every 15 minutes, that means kicking off my users to do so. That's not an option. So in my case it's unusable. But even without it being a server, having to do that every 15 minutes or less... barely usable. And the competition (VMWare Fusion) has issues of its own (that are what made me switch to Parallels 8 months ago in the first place) but their issues don't make it unusable, so that's looking a lot more attractive at this point. Yes, Parallels has some features VMWare Fusion doesn't, that I'd really like to stick with, but if this issue doesn't get resolved soon, I'll be done and I'll make that call for a refund.

To your second question: given that suspending the VM and resuming it resolves the issue, it's hard to know if a technical report can help you, since you only said to send the technical report "If not...". Needless to say I did create one just now anyway, as the issue recurred, in case it might help. The ID is 258542314. The technical report behind that ID was made while the mouse is stuck in the VM and can't be released. I encountered the issue, made the report, suspended the VM, resumed the VM, and then the VM is working again. I then tried to create another technical report immediately, but on trying to submit it, Parallels Desktop crashed ("unexpectedly quit"). I reopened Parallels Desktop again and tried again. Same thing. I now can't send a technical report without it crashing. I suppose I'll have to reinstall the application again. If I have any success after that, I will post more information here.

I sincerely hope you guys can resolve this issue very very soon. Again, I can understand if the issue takes time to fix, but when you're not even communicating anything, meaning you're not even working on it, that's a deal breaker. For an issue that makes the application unusable, I'd think you'd pay a little more attention. If you'll reply to me soon I will work with you on this. If I don't hear anything back soon enough I will call for a refund.

[To others with the same issue, I'd recommend a similar strategy. Please give Dmytri what he's asked for and let's do everything we can at our end to help get this issue resolved. But if they don't respond promptly to that, then after all of us call for a refund, perhaps then they'll listen.]

 

+1 here.....
Virtualizing OS X 10.6.8 (Snow Leopard) Server in order to run three legacy applications. Been complaining about this for eight months too. Was pleased and excited to see that it was addressed in Mac Update 3 Hotfix 2 (13.3.2-43368). But it wasn't. Still the problem persists. I too reported that the bug was not fixed (Parallels #2521593). I too generated a Technical Data Report, ID #263826970 and sent it in on June 29, 2018. No response since then.

 

@TrondB: Version 13.3.2 mentioned in the release notes that the mouse unclick problem has been fixed. Actually, it is much significantly better now, as I can work for 15-30 minutes without mouse freeze. After that, I still have to revert to remote access, which does not freeze for days.

The suspend/resume remedy does not work always work either. Occasionally I need to reboot the guest. Together with the problem that the VM supervisor does not sync the date after wake up, and resets the clock to the wakeup time if you start ntpdate manually for the first 10 minutes after wakeup, the usability is still rather bad.

 

Agreed, the mouse sticks less often since the 13.3.2 Hotfix. Suspend/Resume always seems to fix the issue for me. But come on folks, let's get this fixed once and for all. And while we're at it, can we address that fact that the OS X Guest does not recognize the Caps Lock Key? Or the issue of super fast scrolling in the Guest? These are all serious usability constraints that meaningful reduce the utility value of the product.

 

I switched to VMware Fusion back in May. Glad I did. Parallels is unusable for Mac guests and the vendor apparently has no intention of ever fixing it. Ashame really.

 

TrondB:

Thanks for the follow-up.

Yeah, I've pretty much given up on Parallels. Their claim that one can run a Mac VM using their software is nothing more than intentionally false advertising. The mouse bug makes this use case impossible, or at a minimum, intensely frustrating.

I too noticed that whenever someone raised the issue, the tech support people would misconstrue the issue as being Windows-related, despite the fact that Windows never enters the equation. It's totally bizarre.

I have no particular insight into what's going on at the company, but given the major degradation in their software's quality over the last few years, and the poor level of tech support, my *guess* is that the brainpower behind Parallels left years ago -- and the employees that remain simply have no idea how the software works nor how to modify/fix it. Though disheartening, this sort of endgame isn't terribly uncommon with software providers. The consensus seems to be that with Parallels circling the drain, and no other viable contenders for reliable Mac VM support, the whole "run MacOS in a VM" idea may be an impossible (or at least highly unreliable) use case for the foreseeable future.

 

I can report that there seems to be fewer bugs in older versions of Parallels (which they should be thoroughly ashamed about) - I have Snow Leopard Server guest in Parallels 10 in Mavericks host, and the main bug is that Apple's Preview doesn't work. So I'd recommend you all go to the gates of Parallels Plex and protest loudly, and demand they give you an older version of Parallels as they seem to mess it up each time they update it.

 

Alright... Well I'm giving up.

Spent US$107.51 on this last October. Found this issue straight away, posted about it straight away (I'm the original Poster here). Haven't been able to use the product for more than a few minutes in all that time.

I like Parallels desktop more than VMWare Fusion but I've had to stick with Fusion for this entire time, while hoping they'll fix this issue and I can finally start using Parallels Desktop. But when the support people consistently don't even respond for months at a time, I'm done. At this point, even if they do fix this, I don't want anything to do with a company with such terrible support.

I'm on the phone with Parallels right now seeking a refund. I've pointed him to this thread. He put me on hold for a few minutes while he read it (no problem there of course, I want him to do due diligence). So now he's read it, but is still telling me "I'm sorry our refund policy is we'll only refund within 30 days of purchase."

Not surprising, but I will fight this. Might I recommend others do the same.

Just asked to to speak to his supervisor. He put me on hold for a few moments then came back and said his supervisor will call me back within 30 minutes.

I'll post more as it happens.

 

Three and a half hours later (so... you know... a little more than 30 minutes) and I get the call back. The supervisor is going to speak to "management" and see what he can do.

 

Well... I have been provided a refund as of this morning.

To others posting in this thread who have experienced the same issues as me: it seems they will provide refunds as a result of this issue.

To Parallels support: I like the product and would like to come back to it if this issue is ever completely fixed. I will monitor this thread and if you can report back to claim that it is completely fixed then I would welcome the opportunity to come back to it. In the meantime, thank you for providing the refund for now.

 

Hi All,

New poster here.

Recently I downloaded the Parallels 14 demo to test hosting Snow Leopard Server in Mojave on a 2010 3.33GHz 12-core cheese grater (to maintain a legacy Carbon app in Xcode 3, 'cuz I can't find a Mac-flashed Metal-capable PCIe video card that will boot both). Of course I immediately ran into the (as I subsequently discovered, already infamous) scrolling glitches and sticky mouse bug others have described here, which together rendered the guest OS useless.

Well long story short after some head-scratching I think I've come up with a workaround for the evil sticky mouse button bug that seems 95+% reliable. It requires loading a wee faceless bkgnd app I threw together called mouseUp at guest login, which registers a hotkey (currently F19) that when invoked posts a mouseUp to the guest's global event queue.

Basically it's a thin Cocoa wrapper around deprecated Carbon APIs, because AFAICT writing to the console from Carbon has been broken since sometime around 10.10 (so for the console status messages mouseUp writes to work in recent OS versions it needs to use NSLog) but there's no way to do this with pure Cocoa (at least not without being explicitly assigned permission). As compiled mouseUp should be able to unstick up to three mouse buttons in any version of the Mac OS Parallels can host (all of which apparently evince the same issue). Adding support for more mouse buttons (if someone needs this) would be trivial - it just seemed like overkill.

With mouseUp installed, when your guest mouse sticks (and it will...) just press F19. You should hear a beep and see a status msg written to the console. Then with the Magic Mouse or trackpad (or whatever you're using) try to scroll up or down with a swipe or via the scroll wheel. After this you should once again have control of the stuck mouse button. If not, wait about fifteen seconds and click again.

In my experience this works 90+% of the time on the first click, and most of the remainder on the second. If your mileage varies significantly, please let me know...

I sent this to Parallels a week ago (in hopes it might help them root out the problem) but based on their glacial response to date to the issues ahead of this one in my bug report it could be months before they even get around to looking at it.

So in the meantime I figured I might as well post it here for the benefit of others who might be struggling with the same glitch.

Full source code is included in the zip file - use it however you see fit.

Enjoy!
-Mark