Terminate called after throwing an instance of 'std::system_error'

[roshangm1 (Roosan Gm)]

Hi, I was trying to install and run DeepSpeech ClI on Intel AI DevCloud instance but got this error while starting off.

[u8748@c009 ~]$ deepspeech models/output_graph.pb audio-test.wav models/alphabet.txt
Loading model from file models/output_graph.pb
terminate called after throwing an instance of 'std::system_error'
  what():  Resource temporarily unavailable
Aborted

I feel that this is due to some kind of restrictions in the instance but I am not sure about it and also not sure how to solve this. Please let me know if you have any idea around this.

[lissyx ((slow to reply) [NOT PROVIDING SUPPORT])]

How much memory do you have on this ?

[roshangm1 (Roosan Gm)]

This the output from cat `/proc/meminfo`:

MemTotal:       16267196 kB
MemFree:          157872 kB
MEmAvailable:   14090928 kB

[lissyx ((slow to reply) [NOT PROVIDING SUPPORT])]

So that’s 16GB. It should be okay then. Without more details on your system and the potential limitations, it’s going to be hard to do divination

[roshangm1 (Roosan Gm)]

Hmm okay. If you want, I can provide more details about my system. I am not much familiar with this error. What details you want to know about ?

This is the one I am using right now.

Intel® AI DevCloud - Intel AI

[lissyx ((slow to reply) [NOT PROVIDING SUPPORT])]

Thanks. Any particular reason you try to use this platform? I don’t even know where this std::system_error comes from. Can you gdb into ? Or try with the C++ native client code to check if it’s because of Python ?

[roshangm1 (Roosan Gm)]

By platform, did you mean DevCloud ? If yes, they are providing a kind of powerful instance free for educational purpose. So not having enough powerful device with me, I am using this platform.

Yes, I think, I can gdb into:

GNU gdb (GDB) Red Hat Enterprise Linux 7.6.1-80.el7
Copyright (C) 2013 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
and "show warranty" for details.
This GDB was configured as "x86_64-redhat-linux-gnu".
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>.
(gdb)

I am not sure how to check what is causing the issue. Any link or guide you can provide to test this ?

[lissyx ((slow to reply) [NOT PROVIDING SUPPORT])]

I am not sure how to check what is causing the issue. Any link or guide you can provide to test this ?

No. Again, try to run within gdb and/or try with the C++ native client: https://index.taskcluster.net/v1/task/project.deepspeech.deepspeech.native_client.v0.1.1.cpu/artifacts/public/native_client.tar.xz

[roshangm1 (Roosan Gm)]

Oh okay. Let me try this. I will let you know how it goes.
Thank You!!

[lissyx ((slow to reply) [NOT PROVIDING SUPPORT])]

By platform, did you mean DevCloud ? If yes, they are providing a kind of powerful instance free for educational purpose. So not having enough powerful device with me, I am using this platform.

You don’t need that powerful system for just running inference, and requirements are already going down with v0.2.0 upcoming.

[roshangm1 (Roosan Gm)]

Awesome thing to know about. Actually, my current device is terrible. Thanks for the information though. I will try to do things in the instance for now and later will try in my device also.

[lissyx ((slow to reply) [NOT PROVIDING SUPPORT])]

How much terrible is terrible ?

[roshangm1 (Roosan Gm)]

It’s 2011 Macbook Pro with 4GB RAM. Isn’t it terrible ?

[roshangm1 (Roosan Gm)]

@lissyx BTW, I tried to do this with native client but now got another error instead:

terminate called after throwing an instance of 'St9bad_alloc'
which seems to be a memory issue for me.

Any words on this ?

[roshangm1 (Roosan Gm)]

Memory it was. In the instance, there was restriction on using the memory for user node. What I had to do was, make a job file for the code(task) and schedule it and then only it would use entire memory. It was the issue with the Devcloud solely.

Thanks for your help. You are really nice.

[lissyx ((slow to reply) [NOT PROVIDING SUPPORT])]

Does it have AVX extensions ?

[roshangm1 (Roosan Gm)]

Got this result while doing grep. I actually don’t know about AVX and also about this result. Sorry.

[u8748@c009 ~]$ grep avx /proc/cpuinfo
flags		: fpu de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 syscall nx rdtscp lm constant_tsc rep_good nopl eagerfpu pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx hypervisor lahf_lm 3dnowprefetch invpcid_single fsgsbase bmi1 hle avx2 smep bmi2 erms invpcid rtm rdseed adx smap xsaveopt
flags		: fpu de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 syscall nx rdtscp lm constant_tsc rep_good nopl eagerfpu pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx hypervisor lahf_lm 3dnowprefetch invpcid_single fsgsbase bmi1 hle avx2 smep bmi2 erms invpcid rtm rdseed adx smap xsaveopt
flags		: fpu de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 syscall nx rdtscp lm constant_tsc rep_good nopl eagerfpu pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx hypervisor lahf_lm 3dnowprefetch invpcid_single fsgsbase bmi1 hle avx2 smep bmi2 erms invpcid rtm rdseed adx smap xsaveopt
flags		: fpu de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 syscall nx rdtscp lm constant_tsc rep_good nopl eagerfpu pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx hypervisor lahf_lm 3dnowprefetch invpcid_single fsgsbase bmi1 hle avx2 smep bmi2 erms invpcid rtm rdseed adx smap xsaveopt
flags		: fpu de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 syscall nx rdtscp lm constant_tsc rep_good nopl eagerfpu pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx hypervisor lahf_lm 3dnowprefetch invpcid_single fsgsbase bmi1 hle avx2 smep bmi2 erms invpcid rtm rdseed adx smap xsaveopt
flags		: fpu de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 syscall nx rdtscp lm constant_tsc rep_good nopl eagerfpu pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx hypervisor lahf_lm 3dnowprefetch invpcid_single fsgsbase bmi1 hle avx2 smep bmi2 erms invpcid rtm rdseed adx smap xsaveopt
flags		: fpu de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 syscall nx rdtscp lm constant_tsc rep_good nopl eagerfpu pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx hypervisor lahf_lm 3dnowprefetch invpcid_single fsgsbase bmi1 hle avx2 smep bmi2 erms invpcid rtm rdseed adx smap xsaveopt
flags		: fpu de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 syscall nx rdtscp lm constant_tsc rep_good nopl eagerfpu pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx hypervisor lahf_lm 3dnowprefetch invpcid_single fsgsbase bmi1 hle avx2 smep bmi2 erms invpcid rtm rdseed adx smap xsaveopt

[lissyx ((slow to reply) [NOT PROVIDING SUPPORT])]

Seems like there’s AVX exposed, but you confirmed it was an issue related to the memory

[roshangm1 (Roosan Gm)]

Yeah!! Working okay for now at least.