+36
Started

Portal not responding

Phil K 6 months ago updated by Davison yesterday at 11:30 a.m. 282
ConnectWise Control Version:
19.6
Server Affected:
Host Client Affected:
Guest Client Affected:

Answer

PINNED

As requested, I'll keep Mono-related updates in this post. Updates are posted on Mondays for the previous week.

* July 6: Fix submitted for cert chain issue; preliminary testing of new 20.2 build; verifying compatibility with newest Ubuntu LTS release

* June 29: Continued working to resolve cert chain issue

* June 22: Found and fixed web server restart issue in the course of investigating DB maintenance bug; working on certificate chain issue

* June 15: Continued on DB maintenance bug

* June 8: Wrapping up some other work, so engineering should have more bandwidth; fixed extension loader bug; started work on bug with editing DB maintenance plans

* June 1: Engineering was mostly diverted to general server bug fixes and other tasks; should be able to get back to Mono issues this week

* May 25: Working on general server bug fixes (i.e. not Mono-specific); large rewrite of service life cycle-related code that should address some long-running service restart issues (and hopefully works well running on Mono)

* May 18: Fixed post-setup hang/crash bug; working on extension loader bug

* May 11: Fixing 20.2 for the next release; language version-related issues appear resolved, fixing other known bugs now

* May 4: continuing runtime/compiler update; due to further Mono-breaking application changes in 20.4 (and the possibility of more to come), decided to try and fix issues in the 20.2 release until the breaking updates are complete

* Apr 27: updating Mono fork to resolve issues with newer C# features used in the application; work temporarily suspended mid-week to resolve escaped bugs in 20.3 release (I believe these are all resolved and in testing now); floated suggestion about back-porting fixes (no promises yet)

+1

All our clients have stopped connecting. Services seem to be stopped on windows clients. Event log shows this.

System.Net.Sockets.SocketException (0x80004005): An established connection was aborted by the software in your host machine
at System.Net.Sockets.Socket.Receive(Byte[] buffer, Int32 offset, Int32 size, SocketFlags socketFlags)
at ScreenConnect.SocketNetworkConnection.Receive(Byte[] buffer)
at ScreenConnect.NetworkConnection.OnReadStreamNeedsBufferCycled(Object sender, EventArgs e)
at ScreenConnect.Extensions.RaiseEvent[T](Object sender, EventHandler`1 eventHandler, T eventArgs)
at ScreenConnect.BufferStream.OnNeedsBufferCycled()
at ScreenConnect.BlockBufferReadStream.Read(Byte[] buffer, Int32 offset, Int32 count)
at ScreenConnect.Extensions.ReadByteDefault(Stream stream)
at ScreenConnect.BlockBufferReadStream.ReadByte()
at System.IO.BinaryReader.ReadByte()
at ScreenConnect.MessageSerializer.Deserialize(BinaryReader reader, Type requireBaseClass)
at ScreenConnect.EndPointManager.ReceiveMessage(BinaryReader reader, Type requiredBaseMessageType)
at ScreenConnect.SocketEndPointManager.RunIncomingThread(ThreadSharedState threadSharedState)

Over 200 clients we have lost access to. This is ridiculous. 
Over 6 months without a fix.

I've not got to call every client to get them to restart. 

We are seeing the same thing. Not sure why new problems keep arising when OS state is the same. Very frustrating!

+1

We saw this yesterday as well.  New sessions did connect.  It did seem like overnight many of the sessions did reconnect.  

I think at this point, having Eric post weekly updates is basically useless.  His last update where anything Mono was touched seems to be back in April.  We should be demanding daily updates on this!  Make them post and acknowledge every day that they are ignoring this issue and shirking their responsibility to paying customers.  

Thanks for yet another support outage now going on over 30 minutes!  

+2

Don't you guys think the timeline on this is getting a little embarrassing?  5 months and essentially no progress on the bug.  On top of it, you haven't even shown that you are working on it!  Please give everyone following this thread one good reason why we all shouldn't move to SimpleHelp  (who is currently offering a sizable discount to Control users, and more importantly, a working product)?  

+2

It's more then a bug, they should be rebuilding it from the ground up (doubt it though). I would be embarrassed. Maybe they no longer have Linux developers on staff? Who knows.


I used to only update once a year... now I'm gun shy on that

+1

The result of all of this will be "sorry guys, linux on prem is being depreciated"

+2

I've been assured by many people that's not the case, I pressed them hard on it.  It would seem like that is a lie at the moment.  

+4

The real issue I have is the fact that I'm now having to convince not only myself and coworkers to be patient but our customers too. "Give it a few minutes you'll see it come up soon" "Bare with me I'm waiting for my end to load too..." to the point where "Give me a few seconds" is turning into "Can i give you a call back? I'm not sure when this will be ready to use again". How can I tell my customers "wait just one more month for connectwise to probably still not have fully functioning software!". First your problem became my problem. Now your problem is becoming MY customers problem. SimpleHelp is looking more and more attractive as I wait with zero timeline.

+5

Been following this thread for a while and its quite depressing seeing how little time and effort the linux version gets. I stumbled onto this problem as well earlier when I upgraded to 19.6 which made things really bad. I managed to use an older backup though restoring to 19.4 but clients are still on 19.6 but it haven't affected anything though.

Incase any of you missed the archive and are desperate and wanna try, then make a backup of your current setup and download 19.4 from https://www.connectwise.com/software/control/download-archive. If you can make it work then it will stop hanging and hopefully just work like it used to before.

Regarding focusing so much on Windows sickens me. I really don't wanna run a fully fledged Windows server with all the crap that comes with it. Its easier, cheaper, faster and safer to maintain a linux setup than maintaining a bloated Windows Server install. Its enough doing support for customers on their Windows machine. I don't want to continue working on Windows and sudden changes that takes more time to fix. 

Debian/Ubuntu takes maybe 1-2min to upgrade (including reboot) and is simply

#apt update & apt upgrade -y

#reboot

Done!

Thanks to Eric though for the updates!
I know you know but worth repeating that its not your fault but rather decisions/choices made higher up.

+5

Web Portal unresponsive yet again.  Is there any work being done on this?  Do we have any hope of this getting fixed?  We support our own software here.  If we had a bug this large go untouched for this long, our users would be calling for our heads.  

+5

I got super lucky this morning! My first phone call of the day and I did not have to wait 20 minutes to create a session. By the time I was finished with the session... BOOM! Can't create sessions again. Can't say I was shocked. If my luck continues and the site is working for my next call I think it might be a good idea to play the lottery this week. I think if I were to get 3 calls in a row where I didn't have to wait to create a session I would be assured a winning ticket with my divine luck. 

+1

What if we all start emailing the executive leadership team at ConnectWise? Get this on their radar that they have customers that are suffering with product issues. Keep it nice, keep it short, and get our point across that this is not acceptable to have long standing bugs that are service impacting. Anyone have contact info for someone on the executive team?

https://www.connectwise.com/company/about-us/executive-team

+1

Most engineering teams catch an awful lot of heat when customers start reaching out to their executive team on LinkedIn. Probably a good place to start.

I did that a few weeks ago.  Been in touch with two people on the executive team, but as you can see nothing has happened yet.  I'm keeping my fingers crossed.

+1

It appears I have jinxed my good fortune. Unable to create sessions again for what looks like the 4th or 5th time today already. Unlucky for me that it started happening as I started a phone call this time. This customer was already upset about something not working on their end, which I would have easily been able to get fixed up for them with a help session. Instead I had to tell them "I actually can't help you right now," after I had just said I would. Making someone wait like that, especially when you would be able to easily help them, feels and looks infuriatingly bad on both ends. I have had dozens of phone calls like that in the past few months where I tell someone I can help them and then have to tell them 30 seconds later that I'll have to call them back. It slows their day down, it slows my day down, and makes me look like a POS trying to put them off. It is beyond annoying and is absolutely disgusting that it has been going on for this long. Gross. 

Dear Connect Wise and Forum,

At the Very Least how about a Migration Application or simple instructions so we can Migrate our CONTROL Server back over to a Windows 2019 Server temporarily until this is fixed? Or to the Forum - Has anyone done this or able to share how to do this? Windows has it's own issue's but if Control will run Stable on it I am in. Thanks!

We did this to move from our on premise windows to hosted linux, now looking like we may move back....

+4

This shouldn't even be a thought though.  We all installed on a supported OS, and we all paid our support fees.  Why let them move the goalposts when it should work!

+1

Another day another reboot of our server to alleviate memory usage. Another day, 10.... 15.... 20... more conversations with my clients having to wait on ME for service. Sure, I can verbally walk my younger, more tech savvy clients through what they need help with but what about my older clients I've trained to rely completely on my ability to control their pc? My clients that begin the phone call with "can you get on my computer to see what this is doing? It's acting all squirrely!" now have to wait on my tool acting all squirrely.

+1

@Eric

Would it be an idea to ask devs if customers could downgrade without issue?

I believe the best way to approach this would be to get all back to 19.4 which is the last stable one and then start migrating to the .net core as mentioned earlier which should free up resources long term instead of managing both mono and current Windows version.

I could make a guide (if interested) how to create a fresh new 19.4 install and then it should hopefully work if stopping screenconnect service, copy current App_Data folder over to the fresh 19.4 install and then start screenconnect service again but if devs leave no guarantee then its a risk that you will have other issues perhaps with your current unattended connections.

+2

The minute anyone start talking downgrade to fix an issue that has been going on 5+ months is when we all need to realize we're being strung along

I see your point but I also wanna avoid getting into this place again which is why I repeated the mention of .net core which was discussed earlier. Fact is (as far as we currently know) that nothing remarkably has been done to fix this issue in 5+months and we can only guess the reasons for it. If there are repeated issues getting mono to work from time to time further down the line then we'll surely have other issues again in one way or another. Having a unified system for both Windows and Linux instead of supporting .net core and mono would surely bring less headache for devs and more time to focus on important stuff wouldn't you think?

But if you rather wanna migrate to a Windows server or run a buggy 19.6+ installation thats perfectly fine by me but personally I cant see how I could be productive in any way with how it currently works which is why i'm offering help to those that wants to avoid Windows but still be productive on Linux.

@John123 - Your point is valid, but the issue remains that ConnectWise has completely abandoned their end of the deal here. To downgrade and run an old version is basically validating their lack of effort. 

@Chris

Im not disputing that, however I feel that we need to come up with solutions and the one I mentioned is in the end hopefully a good long term solution. Together with that is also that we should be reimbursed for those wasted months with an added year on our subscription license (as an example) to regain our trust in the company and to show that they do take this seriously instead of trying to sell a new subscription license when this is issue is ongoing (which is kinda awkward and should be better handled)

+3

I'm a bit tired of this issue persisting. I've been checking in for months and echo a lot of what others have said above. 


This is the writing on the wall for ConnectWise Control for us, we've been very patient but 5 months with no resolution is not good enough (not even close). How can we have confidence in a product where critical bug fixes are this slow?

+2

* June 15: Continued on DB maintenance bug

This is shameful.  @Eric - You should get some upper level people involved in this thread.  It's not fair to you to be taking 100% of the heat for this issue.  I'm assuming you're not the one who is making the decision to ignore this issue and leave a good chunk of your paying users without a functioning product.

+2

Ok some how we need to be able to get this resolved. What are our options? Go to the news? Message the guys on Linkedin?


Here are some of the guys in charge:

CEO - https://www.linkedin.com/in/jasonmageecoo/

Chief Product Officer - https://www.linkedin.com/in/jebishop23/


Clearly no one from the company cares about fixing this issue. I've seen faster fixes from Microsoft lol.

Anybody have any contacts at Arstechnica?  I'm sure they'd be interested in writing a blurb.

+1

I'm trying this person: https://www.linkedin.com/in/craigmfulton/ Chief Customer Officer. Will report back if I get anywhere.

I've been in touch with Craig a few times.  He's replied, but I'm not sure how far it gets after him. 

Latest I got from the team...


The next stable release of ConnectWise Control for Linux server has been delayed as the Engineering team works to resolve Mono issues. We’ve been making a lot of really great updates to ConnectWise Control to improve overall performance and stability, but the changes require a large amount of maintenance to our Mono fork. The fact that some of the Mono server issues only become apparent in high-load scenarios adds to the complexity of resolving these issues. While there is still no ETA on the next Linux stable, the team is working quickly to address known issues. We should have an update soon.



Andrew Robinson




So this update and @Eric's update conflict.  Our weekly "updates" document no progress or work on this bug.  Maybe @Eric can shed some more light?

+2

Hi Eric,


Thanks for engaging here - I've read this thread and didn't see the answer elsewhere. Does Connectwise have any plans to remove the "Stable" designation from 19.5 and above? I spun my wheels for a long time on this yesterday and isolated the same problem listed here, to have support say to me "oh yeah, we know about that issue" - is this version actually stable for most customers? I had problems in the recommended environment (Ubuntu 18.04) and the web interface wouldn't even load due to Mono issues - seems different from the issues others reported here, but related. I wouldn't have installed it if it was a prerelease instead and would have appreciated the company's acknowledgement of the issue beforehand to save me time since the end result, regardless, is that I'm running 19.4 (but now have to do it with 20.1 clients). Can those releases have their "Stable" label removed to avoid spreading damage further?

Also, to add a few more details since, again, I'm not having the same exact issue - I don't get random crashes, I get an error even loading the web interface that points to correctly compile mono. I've since restored from backup, but the errors I got were something along the lines

  1. An issue with files in the temp folder where it places its DLL - I think that error said something like "Sharing violation on path" Resources.Default.resources
  2. the other involved a message like mscorlib.dll method not found - it said it had multiple copies of mscorlib.dll and it was choosing one.

Sorry for not having more specifics still, but hopefully that's useful. This is not a high volume server, and reboots of the service and server were not resolving the problem, even temporarily. The web interface was inaccessible entirely and I just got a full error page. Please let me know if this should be a separate bug report - it still points to Mono, so thought I'd join in here for now.

+1

I had the exact same experience as you around 6 weeks ago after attempting to update, same "oh, yeah, we know about this"  response from support suggesting that I restore to 19.4.  It's disappointing that there's no preventative measures stopping this upgrade path just now, I could have saved hours of head scratching and restoring with a little heads up.  Now I sit here eagerly refreshing every Monday morning only to be disappointed whilst my license burns away to nothing.

+1 for removing stable designation from non-stable versions.

+2

Starting my day by trying to help a client move a SQL Server from one PC to another.  Waiting 20 minutes so far to generate a code for the end user.  Awesome feature!

20 minutes?  What are the specs on the machine you are running SC on?

+1

It doesn't matter with the current software, it just locks up. Reboot the system and it restarts the service and makes it happy again

+3

Warren, it wouldn't matter if it was running on a super computer that only alien life forms possess or an etch-a-sketch.  Something is broken and needs to be fixed.  We've been using this service for years, hardware has never been an issue.  

and @ddomains, I know we can do that, but that's not the ideal solution to this problem, and its only temporary until we have to do it again.  It's current state of performance is tarnishing myself and my teams reputation.  Our admin is looking into alternate solutions.  Sad really.  I thought SC was great (when it works).

I'll troll this page until the issue is fixed or I'm banned.

+1

Not ideal, I hate doing it, but I do what I must right now so that I can support clients.

+1

I know.  I hear you loud and clear my (wo)man.  Keep on fighting the good fight.

well i am not having this issue...the fact i have 20.2 might explain the difference then...20.2 was briefly made available as a "stable" release but got pulled..i have a different issue...but i am unable to return to 19.x..:)


i do not know how long this has been available..has anyone tried 19.6 which is labeled stable? https://d1kuyuqowve5id.cloudfront.net/ScreenConnect_19.6.27027.7360_Release.tar.gz

Downgrading as apparently not recommended by screenconnect

+1

I can confirm that the problem(s) started with 19.6 and that 19.4 is the last stable one without issues

20.1 Is what I am running, still major issues

20.1 is what we're on here.  Can (obviously) confirm major issues.

G'day guys,

I reached out to Jeff Bishop on LinkedIn. He took the time to respond to me even though I told him I'm just a small time IT business in Australia. So here goes:


Steven,

I do apologize for the problems you are running into. I was made aware of the issue this morning and I talked to my team. The issue seems to revolve around that fact that we built out our Linux support for hosting the application, many years back, with techniques that just don't seem to be scaling well with some of the required feature changes to support performance and scale that we need implemented. The product chief architect is on it and working with a few developers to find a path forward but it has been a tough one for them to pin down.

We will continue to work on this and my lead PM for this product line will be posting something out to the forum ASAP.

Our intent is to fix this issue but I do want to list out some of the options that are available today in case you were not aware.

1. This issue is 100% around hosting the Control application on a Linux server. If on a Windows OS this isn't an issue. Not sure if that is an option for you but wanted to mention it. As many people have moved to IaaS hosting and we don't use full SQL the cost may not be that substantial for switching. And moving things over is pretty easy.

2. The older release before we started implementing the performance enhancements mostly for scale should also resolve the issue for now. As we finish getting the new version out that should remediate. I believe that release has been posted out. Its pretty easy to test rolling back to an older version and reverting back if any issues. The help site has pretty clear steps on this, and team can help if you don't see those.

3. Some customers are not aware that you can go to our cloud for just a a bit more than what it costs to renew maintenance and host this yourself. Its an option that we only provide to our legacy ScreenConnect customers who have our legacy licensing model.

But again our intent is to fix this issue and we are actively working on it. We seem to fix one thing and break a few more which is where much of the hold up as come. We did have to move some engineers around at the start of COVID to assist partners moving people to work from home, but we then got those engineers back onto this issue ASAP.

I hope this helps a little. I know its not a resolution or a date, but we will keep working on it.

Regards

JB


+2

What concerns me about JB's answer to TechCare is that he "was made aware of the issue this morning and I talked to my team."  I am seriously wondering whether their CEO Geoffery Willison is 'aware' of our issue.  Apparently no one in upper management knew about it until this morning.  How can that be after so many months?


We should follow TechCare's example and reach out to upper management on LinkedIn, then see if the support resources get re-prioritized.  My guess is that Linux on-prem clients are the lowest priority but they will attend to our needs eventually or dump Linux on-prem entirely.

I am sticking with Control on Linux for now since the price is right and the crashes are manageable for my workload as a single tech. However I can't imagine how miserable this must be for a busy IT department with multiple techs and clients calling and expecting support.

I am small-time too, just myself and 65 machines. With 19.6, I reboot twice a day and, luckily, this keeps me mostly stable. My heart also goes out to bigger providers who are not anywhere near as lucky.


I have two grandfathered, ScreenConnect licenses. Last time I did the cost analysis of switching to a SaaS solution, it was double what I am paying now, but perhaps there is a special rate for the SaaS solution that I am unaware of. Does anyone know where I would find that offer?

I do not think anyone is aware of this issue with Upper Management. Case and Point I had a call with Sales yesterday regarding getting licenses for their New Control Extension for users that need to work remotely.  We are running Version 20.1.27036.7360 on Centos 7 Server. The Sales Person asked me to select Admin then License, the applet did not work and responded with a screen stating System.IO.DirectoryNotFoundException Could not find a part of the path "/tmp/root-temp-aspnet-0/348dabd2/uzus5rik.tmp" and it referenced . He then asked me to selecte Extensions as this is where you can add the license to DEMO this no remote access module for home workers and it also responded with System.IO.DirectoryNotFoundException Could not find a part of the path "/tmp/root-temp-aspnet-0/348dabd2/uzus5rik.tmp". Both error Screens stated the following at the bottom of the screen. 
Version Information: 5.18.1 (master/dc6f3b3 Tue Jan 7 18:48:02 UTC 2020); ASP.NET Version: 4.0.30319.42000
Powered by Mono

I told the Sales person that access to ether of these modules has not been working for some time and asked if he was aware of this issue and what the Company and or Programmers (if they have any) has been doing to get this fixed. His response was yeah I heard there was a problem with TLS. I stated that the Problem is with MONO not TLS and there is a long list in the Univiversity Section with many many complaints and no one from the Company has even responded. I requested to let someone know and he said something about that was above his Pay Grade, I then got another Sales Person named Nathan and requested for him to find out and get back to me as I could not possibly consider purchasing any further licenses as the core product has turned unstable and unreliable. He stated he would and email me back. That was yesterday and I still have not received any email or call back on the info I have requested. What really STU's me to boot is I just spend $300 with another vendor to install a PROXY on our Centos Server and set it up to use a certificate from Let's Encrypt. LINUX is so much more Secure with so many less CVE's I want to use it and do not want to go back to WINDOZE.


I will update the Thread and share with all if I do get a reply back from this person on any info I receive.

Sales is aware of one thing.... SALES. they could care less about anything else. This goes for all companies

+1

The Other WACKED thing in the LINUX Version is if you select Version Check it displays the following

Your Version:
20.1.27036.7360
Latest Version:
20.6.28797.7465
Latest Eligible Version:
20.7.28837.7465
Download Location:
https://www.screenconnect.com/Download

Why in the World is the LINUX Version displaying there is Stable Version to install when there is not.LINUX Control should be checking for LINUX CONTROL Updates and WINDOWS CONTROL should be checking for WINDOWS Updates. This is providing completely WRONG Information and should be fixed and Updated TODAY. Simply having 2 Sections in the Version Check would fix it and take less than 5 minutes. One Section for WINDOWS and one Section for LINUX. Seems like some people at ConnectWise are not asleep at the wheel and not doing their jobs.

19.4 was fine, 19.6 had issues.  I rolled back to 19.4, and then there were some 20.x versions earlier in the year that were awful with tons of issues, but I am on 20.2 that was briefly on the downloads page and it's decently stable.  Some slight delays to get in sometimes (150 unattended) and every now and then a memory issue (1 GB of memory) but overall it's stable.  I restart when updates require it, I was doing it daily until one day a lot of unattended clients never reconnected after the reboot and I had to reinstall on the affected machines.


I'm on 20.2.27296.7376.  Can a staff member make that available for those on 19.x that want to try it out?  I have a download of it if not.


I can see the writing on the wall, Linux is going to go the way of macOS for hosting and it will be Windows only soon enough.  Waiting for things to die down here before I do the migration.

!Gb of ram simply isn't enough even for Linux.  Linux itself needs 1-2 gigs of ram just for operations.  With 150 machines I would say you need no less than 8 gigs of ram...and based on my usage I would recommend no less than 16 gigs.  Here's my system stats:

2x  L5420 @ 2.50GHz, 8 cores

16 gigs ram

2 x 500 GB HDD in MDRAID 1

I have 109 machines under management and SC is using 5.2 gigs of ram itself.  If your machine isn't above 50% swap usage i will be impressed...I put SC on this machine after my 2 GB vm I had it on was at 100% swap and SC started getting OOM killed..SC starting running much better once it had room to stretch...


+2

I really disagree with this - I've run screenconnect successfully and with good performance on DigitalOcean with 512MB RAM and now 1GB of RAM for a long time. It performed very well even with many access clients (but only 1 concurrent session). Certainly scenarios will differ based upon usage (we use more resources for another install with ~100 clients, but not much more), but ScreenConnect has done fine with very few resources for a long time.

What version are you on?  19.6 and some of the 20.x versions have had issues where the service uses a ton of memory until mono throws an out of memory issue and usually only a reboot will fix it.  I haven't had the issue on 19.4 or 20.2.

Nick, my experience is very similar to yours: DigitalOcean, previously with 512MB, now with 1GB, 65 hosts, two concurrent connections. It's not blazing fast, but it is very serviceable. v19.6. (I could have reverted to a weekly backup with 19.4 on it, but I didn't experience the problem enough for over a month, so that backup went *poof*--rats!)

Sorry to hear that! Fortunately (unfortunately?) my upgrade experience was dire enough that my only option was to rollback immediately, but one thing I'm taking away from this too is that I'll want to keep a snapshot of my last known good version in addition to addition to my backups in the future. I haven't done that for this service in the past, but will now.


As an addendum, I just looked, and on my 1GB RAM server, the whole server is using less than 50% RAM and isn't touching the disk. There were a few versions many years ago where after about a month of uptime, SC would peg the CPU, but I haven't had that issue in a long time - just the new ones post-19.4

+3

One thing that should be obvious by now but is still worth mentioning:  Everyone should disable auto-upgrading the client agents before any host upgrade.  Test out the new version for quite some time with only a small number of client agents manually upgraded before rolling out the update to all clients (this shouldn't prevent still being able to work with clients on an older version).


While clients using a version higher than the host (in the case of rolling back to an earlier version) *may* still connect there isn't any solid guidance from CW with this scenario (and will undoubtedly not work with certain protocol changes).


It would be a really good feature if CC would standardize on a baseline protocol that was enough to allow compatibility between versions so that the host could push a reinstall to a client even if the client was running a newer version.

@Eric - So, correct me if I'm wrong, but it looks like there was still nothing done to address this particular bug?  

I'm reading back through the updates and i'm actually not sure when the last week they actually were working on the Mono issues? Maybe between May 11-18? If i weren't suffering this bug i'd actually be kind of impressed that Eric keeps providing updates like this knowing they don't help us at all.

Pretty sure it was April 27th. You would think they'd be embarrassed by these "updates". 

This is disappointing. At this point in time I feel it's asinine to continue a relationship with ConnectWise. 

+1

We should really all request our Support charges back from January.  We've literally gotten no support.

+1

Agreed. The email I received from support basically disavowed them from everything:

"As mentioned before we do support Linux but have been struggling to get past the mono issues. We do not guarantee perfection, we do not guarantee zero downtime, but we certainly try to do what we can to prevent issues from happening."

I've thrown in the towel; CW wins. I'm moving to Windows Server running in Azure, B2ms instance size which is similar in specs to the Linux VM that I am currently running. Using ActionPack credits to cover the expected cost, which might be 3x the amount that I am currently paying ($60+ monthly vs $20.00 monthly). Add in a few more bucks for RMM monitoring, software patching, and AV since it is Windows. This migration better go smoothly or I'm going to be even more annoyed than I currently am.

We are doing the same.

+1

No! Don't give in! This MUST be resolved. We moved to Llinux with their blessing that everything would work. For me to move back to Windows they would literally have to write me a cheque for $25,000 for all this headache.

Sorry, my business is too important right now, Moving it to in in house NUC windows box for now, when things are stable we will move back.

At the end of the day my clients are most important, and this is the right move for my clients.

I hear you! Personally I would be more included to change solutions rather than switch back to Windows. Connectwise has lost all my trust.

It was always very stable for us on windows when we had it hosted in our office on a 75/75 connection, When we moved 2 years ago FiOS was no longer an option, so I threw it on a hosted linux server fearing 250/10 would be an issue (upload wise). We now have 1g/1g service at one of our locations, so this is kinda a no brainier for us at the moment until they get their act together again.

If you think this is bad, There's a 3 year old open bug about Samsung tablets not working properly.. 

I think it's time to say goodbye to these guys. 

It's somewhat frustrating, paid support period goes by and "nothing" happens, tech scammers are taking the hosted route, but I wanna stick with my on-premise linux deployment. Tech Scammers using ConnectControl

Just started watching the video, their hosted portal is still up.

@Eric - Are we just not getting an update today? 

Where is today's update?

+6

Here's my guess:

* June 29 - Happily worked on many small Windows Server issues.  Investigated some typos in UI.  

+3

All kidding aside, is there actually no update today?!

+2

Hellooo? Still no update?

+5

June 29th update was just posted, and still no work on this issue.  ConnectWise should really be ashamed at this point.  This is just atrocious.  

+3

So like there's totally going to be another update this week right? That June 29th one that was posted on June 30th was quite the let down! Kind of crazy that the last time the ACTUAL problem in this thread was even mentioned was in the first update on April 27th. Is there literally NOTHING to say about it? Would definitely lead me to believe that nothing is being worked on for a major issue which seems pretty weird! 

+3

I think all of us here need to demand answers and a timeline.  What work has been done, what is left to be done, and when will this fix be complete?  We've all paid for support and updates, yet those updates and support have been non-existent.  If Eric can't answer these questions, get somebody in this thread who can.  If the questions can't be answered here, what support channel do we need to use?  

+1

I think we should be receiving DAILY updates on this issue. 

I have a user on the phone right now that I can't help because it just decided right now would be a good time to stop working. 

I upgraded from v5.6 to v20.1 and found that the server would often crash after a support session, requiring multiple restarts a day. I thought this might be due to increased RAM requirements but after reaching out to Connectwise support they said it was due a memory leak with Mono. To their credit they sent me straight to this thread.

Yesterday I rented a Windows VPS just for CC, with a hope to return to the Linux version as soon as possible. A downgrade of the server on Linux would have been better but interrupting customers to downgrade the Windows client would be a bit embarassing.

+2

It's nice that works for you. Unfortunately several of us have security compliance profiles that require the sessions to be on internal infrastructure and utilizing specific Linux distributions. disConnectwise seems to be "waiting them out" in this case, assuming all these customers will find other remote support products.

+3

No update yet?  Guessing the issue is resolved.  Looking forward to the new release.

+4

July 6: stuff, things.

+1

Waiting patiently for this morning's download....

+5

I haven't commented on this but was having the same issue after an upgrade in March.  I contacted Support in March and opened up a ticket as I had additional issues including the portal not responding after the upgrade to 20.  Thankfully, I snapshotted the VM prior to the upgrade and was able to revert after keeping the support ticket for a week without resolution.  I was contacted by their sales department when my license renewal came up in June and gave them an earful, as I refused to pay for the renewal and I was also needing to add additional concurrent sessions, so they have at least lost the revenue from my sales.  They told me to just switch it over to a Windows server or move to their cloud offering.  If you can't support the current system, how can I trust you to support your cloud, or a Windows based system for that matter? I agree that this is completely ridiculous and reflects very poorly on how Connectwise handles their customers.  I'd like to think that if I had an issue that impacted this many of my customers and could not handle it in a timely manner, I would probably be out of business.  To me, this has become an integrity issue.  I believe in delivering what you promise.  We pay for support that is non-existent.  This is not acceptable and reflects poorly on your (Lack of) leadership, management, and all the way down to the individual software developers.

You have lost another customer by your inability to deliver a fix in a timely fashion.  I'm moving on to something else and I'll make sure I  never recommend your products or services again.

+2

J Simmons, Please let me know what you find as an alternate as I also have grave concerns on ConnectWise lack of taking care of their Customers properly. I too am not ready to move to their Cloud if they cannot support the LINUX Version and going to Windows opens up a whole other can of worms regarding daily security updates and all the other lacking security features and CVE's that Windows has problems with. This was my point of putting it on a LINUX server in the first place. Thanks

+5

@Brent - If you haven't tried Meshcentral, you should.  We have it running in tandem at the moment for the times Control decides to die on us.  Open source, free, light on resources, and has a super responsive dev team.

+3

Hi Chris - Thanks!! I will take that Car for a Spin!! 

+4

@Eric - So how long before we see a beta?  Or even something related to this in the 20.2 output stream?  (https://control.product.connectwise.com/communities/26/topics/2831-connectwise-control-202)

+4

Any updates?   Linux users who have the software installed have not been working since 19.4.  The 20.1 release is not stable at all.  Are there any plans to actually fix this or should I basically stop renewing support for a product that has no support. 

+3

It would be great if we go slightly more info than Eric's Monday updates.  I think we all want to know when we're going to see something and would feel better about progress if something was shown in the out put stream.

+1

@Eric - So with your latest update that would mean that the problems are fixed? We are expecting a beta version to 20.2 any moment now? Can you provide any further information please?

+1

4% of the Connectwise workforce was let go 2 days ago. Curious if any of those positions were dev positions.

https://www.reddit.com/r/msp/comments/hntjvt/an_inside_view_of_connectwises_reduction_in_force/

+1

@Eric - There is still nothing regarding this change in the Output Stream (https://control.product.connectwise.com/communities/26/topics/2831-connectwise-control-202).  Can we please get an ETA on this fix??

+3

I'm curious about the cadence of releases this year.  On the surface it would seem that CW has a rapid iteration pattern in development as of late, but I'm wondering if a lot of the new point releases are really just very minor enhancements that aren't actually worthy of a full point release.  If they really are spending so much development time on the product then it is clearly time to put the brakes on "new" features for a while and start actually fixing the Mono build that has been semi-broken for the past 8 months and non-existent for the past 6 months.  I'd give them some leeway if they announced a plan for migrating to a newer .NET stack, but the silence has been deafening.