Gizmo Activation on Linux

[oxff]

Just installed the SR22 Entegra Series today. Installed just like the TBM which was nice. When I started XP and loaded into the aircraft, nothing worked and I'm getting a thunking sound backed up with either a mooing or a "oh ohhhh" sound. It's a little hard to tell.

 

I went thru my Log and found quite a few errors. A snip of is below.

dlerror:/home/tim/X-Plane11/Aircraft/X-Aviation/SR22 Series Entegra/TorqueSim SR22TN Entegra/plugins/OpenGPWS/lin_x64/OpenGPWS.xpl: undefined symbol: sio_getpar
dlerror:/home/tim/X-Plane11/Aircraft/X-Aviation/SR22 Series Entegra/TorqueSim SR22TN Entegra/plugins/sr_systavi/lin_x64/sr_systavi.xpl: undefined symbol: ASN1_ANY_it
0:01:25.319 E/SOUN: The sound event event:/cb/cbPush has an invalid command trigger afm/sr/cmd/cb/r3c08
0:01:25.319 E/SOUN: snd error in file: Aircraft/X-Aviation/SR22 Series Entegra/TorqueSim SR22TN Entegra/fmod/SR22TN_AVIDYNE_DFC90.snd line 2482
0:01:25.319 E/SOUN: The sound event event:/cb/cbPush has an invalid command trigger afm/sr/cmd/cb/r3c09
0:01:25.319 E/SOUN: snd error in file: Aircraft/X-Aviation/SR22 Series Entegra/TorqueSim SR22TN Entegra/fmod/SR22TN_AVIDYNE_DFC90.snd line 2488
0:01:25.319 E/SOUN: The sound event event:/cb/cbPush has an invalid command trigger afm/sr/cmd/cb/r3c10
0:01:25.319 E/SOUN: snd error in file: Aircraft/X-Aviation/SR22 Series Entegra/TorqueSim SR22TN Entegra/fmod/SR22TN_AVIDYNE_DFC90.snd line 2494

 

This brought me to the forums here and after reading the "Switches don't work" thread, and realizing what I'm seeing looks like a failed activation. I went looking for Gizmo. And I did indeed find a Gizmo error.
dlerror:/home/tim/X-Plane11/Resources/plugins/Gizmo64.plugin/lin_x64/Gizmo64.plugin.xpl: invalid ELF header

It's modified date under plugins, including the directories under Gizmo64.plugin are from today, however the big notable difference is when I look at the XPLs for Windows and Mac. They're dated 3/15/2021 whereas the Linux plugin is dated 2/25/2018.

 

Gizmo's website says there is no Linux version and there's a bug report that's been open since 2017 on Github. I'm curious how this is supposed to work on Linux or if the Linux support indicator on X-Aviation was a mistake?

[Coop]

This seems to be a library dependency issue with some versions of Linux that we are working on

[oxff]

What can I do to assist you with the troubleshooting and fix?

[Coop]

We are still working through this one on what dependency is needed. We are targeting getting a fix for 1.0.2 at this time.

We have an already finalized patch (1.0.1) that should be releasing soon, this will not have a change on the dependencies with Linux as we are still working on that fix. 

[javaes]

Hi, the same error is happening on 1.0.2 as well. Which Linux distro have you tested? I would really love to finally fly this bird.

[Coop]

2 hours ago, javaes said:

Hi, the same error is happening on 1.0.2 as well. Which Linux distro have you tested? I would really love to finally fly this bird.

Thanks for reporting back. We are looking into it, hopefully we will be able to generate a 1.0.2b with some other changes. I'll PM you when we have something to test. Compiled on Ubuntu 18.04 LTS

[LinuxFan]

I'm also running 1.02 on Ubuntu 21.04. Airplane installs OK but when I start it, no activation is requested and the airplane does not function.

If I can help in anyway ....

 

[Coop]

Gizmo isn't used on Linux. If you can send your Log.txt file, that should help narrow down what is being failed to load.

[LinuxFan]

Log file attached. Please let me know if I can help in ANY other way.

 

Log.txt

[LinuxFan]

If there is a library dependency in Linux, please let me know which Ubuntu library package to install.

Thank you for your help!

 

[LinuxFan]

Coop, any updates?

[Coop]

16 hours ago, LinuxFan said:

Coop, any updates?

Just PMd a build to try

[tinroof]

Hi,

I am another Linux user drawn in to purchasing the SR22-Entegra by the Linux support indicator on X-Aviation's product information. I immediately recognized my error when the installation only supported Mac/Windows installers. I immediately sent an email to support requesting that there be a refund or credit made to the account. This was December 1st 2021. It is now only a week from Christmas and have received no further information regarding how this is to be resolved. More surprising is the fact X-Aviation still indicates the SR-22 Entegra is supported on Linux.

My preferred option would be to have the SR22 functional on my Linux installation of X-Plane. The means to achieving this is to have the gizmo64.plugin linux compatible. It might have been many distributions in the past, but there has been nothing since 2018. For this to occur Ben Russell the Gizmo developer would need to be on board and indications are this is not likely to occur. I hope I have viewed the wrong threads regarding this but the age of this unresolved post seems to indicate otherwise. Shame. There are quite a few products in the market place that use Gizmo. The flight factor A320 comes to mind.

In the mean time, what can I expect from X-Aviation with regards to my non-functional purchase?

Russell Diehl

VATPAC VFR Coordinator

[Cameron]

8 hours ago, tinroof said:

More surprising is the fact X-Aviation still indicates the SR-22 Entegra is supported on Linux.

Because it is.

 

8 hours ago, tinroof said:

The means to achieving this is to have the gizmo64.plugin linux compatible.

That is not correct. As has already been indicated in this very thread above, the Entegra does not use Gizmo on Linux.

 

8 hours ago, tinroof said:

There are quite a few products in the market place that use Gizmo. The flight factor A320 comes to mind.

The A320 does not use Gizmo.

 

8 hours ago, tinroof said:

In the mean time, what can I expect from X-Aviation with regards to my non-functional purchase.

The lack of Linux installer in the latest update was an oversight that will be corrected. In the meantime you can run WINE to use the Windows installer on Linux. All necessary Linux files are in the installed package.

[Ben Russell]

Current Linux market share: 1.27%

[oxff]

Figured since I started this thread almost a year ago now I should post an update.

A few minutes ago I installed 1.0.3. Tho there was no Linux installer, the Windows one worked just fine in Wine (minus some small graphical issues) so no problem.

It would appear the issue I reported is still happening and the aircraft is still not flyable. Click anything and you get a generic "ohooooooh" sound.

So here's my question. Is there a working (or will be a working) Linux build? I'm really trying to not come in here with negativity but it's hard when I spent $60 on aircraft marketed as having Linux compatibility but then not being able to use it due to bugs and a year later seeing the same issue.

If Linux support is still being fixed then no problem. But if this has fallen off the table and won't really be worked on, I would ask that the Linux support be removed from the item listing and I'll submit for a refund. Hopefully if I have to go down the refund route I won't have issues like the previous poster did, I would hate for this to turn into false advertising.

BUT like I said previously. I'm trying to not be negative and still have hope this will work.

[Coop]

In order to debug what is going on, we will need you to send us a copy of the X-Plane log.txt file. You can find a guide to get this file here: 

How to find the X-Plane log.txt file?

[oxff]

16 hours ago, Coop said:

In order to debug what is going on, we will need you to send us a copy of the X-Plane log.txt file. You can find a guide to get this file here: 

How to find the X-Plane log.txt file?

Thanks Coop for coming back and luckily I'm familiar with the log.

Looking thru it, it's the same errors reported in my first post in this thread but since you asked, here is the log.

I'm curious where we go from here? The fix for these errors were supposed to be in 1.0.2, then in 1.0.2b, and now we're at 1.0.3 with the same issue.

sr22Log20220206.txt

[Cameron]

1 hour ago, oxff said:

The fix for these errors were supposed to be in 1.0.2, then in 1.0.2b, and now we're at 1.0.3 with the same issue.

If you look through the release notes, the following is present in 1.0.3:

Disable obstacle database on Linux builds to prevent library loading issues

To assert that the team just dismissed or could care less is incorrect, and I suggest you take a different tone here.

I find Linux a sticky issue as it is. There are so many flavors of it, that it's almost at your own risk without using Ubuntu 18.04 LTS. The argument I commonly get amongst Linux users when we don't support the platform for products is that they understand there are inherent risks due to the many flavors and libraries. Then, when we bring back the opportunity to use our software on Linux, stuff like this happens and it creates a compounding headache all over again.

Give @Coop the opportunity to help you best he can before throwing accusations around of not doing anything about this stuff. He means well.

Thank you.

[Coop]

1 hour ago, oxff said:

Thanks Coop for coming back and luckily I'm familiar with the log.

Looking thru it, it's the same errors reported in my first post in this thread but since you asked, here is the log.

I'm curious where we go from here? The fix for these errors were supposed to be in 1.0.2, then in 1.0.2b, and now we're at 1.0.3 with the same issue.

sr22Log20220206.txt 39.39 kB · 0 downloads

Try this out in the Entegra /plugins/sr_systavi/lin_x64. Regarding OpenGPWS not loading, that one would most likely be some sort of missing dependency on your system as it a third party plugin by Totoritko of HotStart. If you have the TBM or CL650, you can try copying over the OpenGPWS plugin from there, or you can try a version from here: https://github.com/skiselkov/opengpws/releases 

If this doesn't work, please send your full log.txt as there are crucial bits of other information besides the intial loading error that are being left out.

1559314776_Entegra1.0.3bLinuxsr_systavi.xpl.zip

[oxff]

4 hours ago, Cameron said:

To assert that the team just dismissed or could care less is incorrect, and I suggest you take a different tone here.

I would suggest you reread this thread because I've done no assertion of what you claim I have. You are putting words in my mouth and that is not appreciated. The only one coming into this thread asserting any attitude is you. My questions are completely legitimate regarding a product I purchased that is marked as having Linux support and hasn't worked since release (almost a year).

On 2/5/2022 at 1:16 PM, oxff said:

If Linux support is still being fixed then no problem.

I would like to point out this line in particular. I understand Linux can be and is a pain and I understand business objectives change. I would say I've been extremely patient, so asking if Linux support is still on the horizon is not a far fetched question given it's been a year.

4 hours ago, Cameron said:

The argument I commonly get amongst Linux users when we don't support the platform for products is that they understand there are inherent risks due to the many flavors and libraries. Then, when we bring back the opportunity to use our software on Linux, stuff like this happens and it creates a compounding headache all over again.

This attitude wouldn't fly in any sales market. Imagine the outrage this would generate if this was applied to a non-software product. If you bought a car that advertised left and right turns, but then the steering wheel would not turn left, there would be problems. I work with vendors on a daily basis to get software running on Linux, it's only hard when you come into it with assumptions based on how Windows works. Such as building to a dot release specific version of software (just statically link), hard coding paths, or playing games with capitalization. Being Linux is the #1 server OS, it's obviously not too hard to support otherwise these large legacy IT enterprises would stick with a Unix like Solaris or AIX.

4 hours ago, Cameron said:

Give @Coop the opportunity to help you best he can before throwing accusations around of not doing anything about this stuff. He means well.

I am doing exactly this and have said nothing about Coop so again to assert that I am doing otherwise is putting words in my mouth and conflating my intentions with your assumptions.

I'm sad to see I've stirred up your saltiness but I hope you have a better day.

 

4 hours ago, Coop said:

Try this out in the Entegra /plugins/sr_systavi/lin_x64. Regarding OpenGPWS not loading, that one would most likely be some sort of missing dependency on your system as it a third party plugin by Totoritko of HotStart. If you have the TBM or CL650, you can try copying over the OpenGPWS plugin from there, or you can try a version from here: https://github.com/skiselkov/opengpws/releases 

Thank you! I placed the file and loaded the plane (before coping over OpenGPWS) and XP crashed. I then copied over OpenGPWS from the TBM900 and tested again. This time the aircraft loaded and is thus far functional. I will run it through some testing and report back.

I took a look at the OpenGPWS.xpl and it was compiled for System V instead of Linux. This should be an easy argument change depending on how you're building/sourcing the software.

Shipped with the SR22:
$ readelf -h OpenGPWS.xpl
ELF Header:
  Magic:   7f 45 4c 46 02 01 01 00 00 00 00 00 00 00 00 00
  Class:                             ELF64
  Data:                              2's complement, little endian
  Version:                           1 (current)
  OS/ABI:                            UNIX - System V
  ABI Version:                       0
  Type:                              DYN (Shared object file)
  Machine:                           Advanced Micro Devices X86-64
  Version:                           0x1
  Entry point address:               0x59800
  Start of program headers:          64 (bytes into file)
  Start of section headers:          19413912 (bytes into file)
  Flags:                             0x0
  Size of this header:               64 (bytes)
  Size of program headers:           56 (bytes)
  Number of program headers:         8
  Size of section headers:           64 (bytes)
  Number of section headers:         36
  Section header string table index: 35

TBM900:
$ readelf -h OpenGPWS.xpl
ELF Header:
  Magic:   7f 45 4c 46 02 01 01 03 00 00 00 00 00 00 00 00
  Class:                             ELF64
  Data:                              2's complement, little endian
  Version:                           1 (current)
  OS/ABI:                            UNIX - GNU
  ABI Version:                       0
  Type:                              DYN (Shared object file)
  Machine:                           Advanced Micro Devices X86-64
  Version:                           0x1
  Entry point address:               0x6b000
  Start of program headers:          64 (bytes into file)
  Start of section headers:          33307976 (bytes into file)
  Flags:                             0x0
  Size of this header:               64 (bytes)
  Size of program headers:           56 (bytes)
  Number of program headers:         8
  Size of section headers:           64 (bytes)
  Number of section headers:         38
  Section header string table index: 37

 

ps. Please don't take Cameron's representation of me as who I really am. I will help you debug until I'm blue in the face and I understand that issues occur and business objectives change. I'm not concerned with any of that as I understand how things go. I've been a Unix engineer in large enterprises for the past decade so I've seen my share of this.

Edited February 7, 2022 by oxff

[Ben Russell]

Please don't compare software to Cars.

You can still make a left turn. You just have to hold the steering wheel to the right for a longer period of time.

This is amusingly almost an identical experience to Linux vs other operating systems.

You can still get it to work. You just have to do stupid annoying and seemingly un-necessary things to get to your objective.

 

You will also find a reference to Ubuntu 18.04 LTS earlier in the thread.

 

[Cameron]

2 hours ago, oxff said:

I would suggest you reread this thread because I've done no assertion of what you claim I have. You are putting words in my mouth and that is not appreciated.

I am not putting words into your mouth. You are putting words into ours. You literally said, "The fix for these errors were supposed to be in 1.0.2, then in 1.0.2b, and now we're at 1.0.3 with the same issue."

That literally implies we took no action to this point, which is incorrect. You are the first person to report this since the release of 1.03.

2 hours ago, oxff said:

I would say I've been extremely patient, so asking if Linux support is still on the horizon is not a far fetched question given it's been a year.

This is just really lame. I'm sorry, but it is. It's been supported since the day of release. The release was tested with Ubuntu 18.04 LTS, and it works out of the box with such. You are clearly not running that, but that doesn't mean we don't support Linux. That's also putting words in our mouth.

2 hours ago, oxff said:

This attitude wouldn't fly in any sales market. Imagine the outrage this would generate if this was applied to a non-software product. If you bought a car that advertised left and right turns, but then the steering wheel would not turn left, there would be problems. I work with vendors on a daily basis to get software running on Linux, it's only hard when you come into it with assumptions based on how Windows works. Such as building to a dot release specific version of software (just statically link), hard coding paths, or playing games with capitalization. Being Linux is the #1 server OS, it's obviously not too hard to support otherwise these large legacy IT enterprises would stick with a Unix like Solaris or AIX.

I really, honestly, do not care to compare to another industry in any form. We are talking about a niche in a niche in a niche here. X-Plane itself is a small industry, then you have Linux, which is less than 2% of X-Plane users, and then you finally have you, a Linux user using a build of Linux other than the most popular variant available for which the product was compiled and tested on.

Linux excels at servers. I don't dispute that in any form.

We don't need to continue this longwinded drama, so please don't do so.

@Coop will update here appropriately. Thank you for your patience.

 

[Coop]

2 hours ago, oxff said:

Thank you! I placed the file and loaded the plane (before coping over OpenGPWS) and XP crashed. I then copied over OpenGPWS from the TBM900 and tested again. This time the aircraft loaded and is thus far functional. I will run it through some testing and report back.

I took a look at the OpenGPWS.xpl and it was compiled for System V instead of Linux. This should be an easy argument change depending on how you're building/sourcing the software.

This bit is interesting, not sure where the wires got crossed to trigger that -- one of our debugging steps from the beginning has been the copying over the TBM900 and I don't think anyones realized this explicitly till now. These things are hard to get picked up on our end for Linux given all the variations (I can probably count our Linux SR2X users on 1 hand). I'll put this on my list to look at what happened with that shipping Linux build of OpenGPWS.

[oxff]

Thank you Coop! Let me know if you need anything from me or want me to test something. I ran thru the majority of the systems and the only issue is the transponder. Datarefs show it's operational, it's just not displaying. Adding my log just in case.

Log.txt