platform_shared_memory_region_posix.cc error

[trum994]

I'm running into this problem when using the nfcore nanoseq workflow which calls nanoplot. How can I help debug?

2021-12-16 10:48:06,816 No static plots are saved due to some kaleido problem:
2021-12-16 10:48:06,817 Failed to start Kaleido subprocess. Error stream:

[1216/104806.675072:WARNING:resource_bundle.cc(431)] locale_file_path.empty() for locale
[1216/104806.687812:WARNING:resource_bundle.cc(431)] locale_file_path.empty() for locale
[1216/104806.707433:WARNING:resource_bundle.cc(431)] locale_file_path.empty() for locale
[1216/104806.708556:WARNING:discardable_shared_memory_manager.cc(194)] Less than 64MB of free space in temporary directory for shared memory files: 0
[1216/104806.727847:ERROR:platform_shared_memory_region_posix.cc(250)] Creating shared memory in /local/25544375/.org.chromium.Chromium.R3O5XA failed: No such file or directory (2)
[1216/104806.727896:ERROR:platform_shared_memory_region_posix.cc(253)] Unable to access(W_OK|X_OK) /local/25544375: No such file or directory (2)
Received signal 6
#0 0x55555a37cd79 base::debug::CollectStackTrace()
#1 0x55555a2fa633 base::debug::StackTrace::StackTrace()
#2 0x55555a37c95b base::debug::(anonymous namespace)::StackDumpSignalHandler()
#3 0x2aaaaaaf1730 (/lib/x86_64-linux-gnu/libpthread-2.28.so+0x1272f)
#4 0x2aaaaaeb17bb gsignal
#5 0x2aaaaae9c535 abort
#6 0x55555a32830a base::internal::OnNoMemoryInternal()
#7 0x55555a328329 base::(anonymous namespace)::OnNoMemory()
#8 0x55555a328319 base::TerminateBecauseOutOfMemory()
#9 0x55555a3120ab base::FieldTrialList::InstantiateFieldTrialAllocatorIfNeeded()
#10 0x55555a312239 base::FieldTrialList::CopyFieldTrialStateToFlags()
#11 0x555558e6df82 content::GpuProcessHost::LaunchGpuProcess()
#12 0x555558e6c910 content::GpuProcessHost::Init()
#13 0x555558e6c6c2 content::GpuProcessHost::Get()
#14 0x555559289b6e base::internal::Invoker<>::RunOnce()
#15 0x55555a340306 base::TaskAnnotator::RunTask()
#16 0x55555a351cf6 base::sequence_manager::internal::ThreadControllerWithMessagePumpImpl::DoWorkImpl()
#17 0x55555a3519ea base::sequence_manager::internal::ThreadControllerWithMessagePumpImpl::DoWork()
#18 0x55555a39e899 base::MessagePumpLibevent::Run()
#19 0x55555a35259b base::sequence_manager::internal::ThreadControllerWithMessagePumpImpl::Run()
#20 0x55555a32b9bd base::RunLoop::Run()
#21 0x555558d13f18 content::BrowserProcessSubThread::IOThreadRun()
#22 0x55555a36a874 base::Thread::ThreadMain()
#23 0x55555a38ccaa base::(anonymous namespace)::ThreadFunc()
#24 0x2aaaaaae6fa3 start_thread
#25 0x2aaaaaf734cf clone
  r8: 0000000000000000  r9: 00002aaaab9fcf50 r10: 0000000000000008 r11: 0000000000000246
 r12: 0000555556d5e780 r13: 0000555556ca0eb0 r14: 00001de88bd39c00 r15: 00001de88be0a380
  di: 0000000000000002  si: 00002aaaab9fcf50  bp: 00002aaaab9fd1b0  bx: 0000000000000006
  dx: 0000000000000000  ax: 0000000000000000  cx: ffffffffffffffff  sp: 00002aaaab9fcf50
  ip: 00002aaaaaeb17bb efl: 0000000000000246 cgf: aaaa000000000033 erf: 0000000000000000
 trp: 0000000000000000 msk: 0000000000000000 cr2: 0000000000000000
[end of stack trace]
Calling _exit(1). Core file will not be generated.

[Wouter De Coster]

Hi,

I will take longer to respond to issues as I am on paternity leave. This error comes from kaleido, which is a nice way to write the dynamic HTML plots into static PNGs. Unfortunately, it also leads to rather cryptic errors (especially on WSL, it seems). One way to solve this is that I have a try-except around calling kaleido, and this catches some errors without crashing NanoPlot.
To "work around" other issues I recently added a --no_static option to skip the kaleido step, but that leaves you with only the HTML files and report. That probably is not a solution here?

So I understood that NanoPlot did not actually crash, but the expected png was not created?

Wouter

[trum994]

Thanks for the response Wouter. I think you are correct that nanoplot didn't actually crash. Exit status was zero, and the output contained the html reports. However nfcore nanoseq workflow expects the png files which were not created:

Error executing process > 'NFCORE_NANOSEQ:NANOSEQ:QCBASECALL_PYCOQC_NANOPLOT:NANOPLOT (sample1)'

Caused by:
Missing output file(s) summary/*.png expected by process NFCORE_NANOSEQ:NANOSEQ:QCBASECALL_PYCOQC_NANOPLOT:NANOPLOT (sample1)

So I either need to find a way to fix this kaleido problem or tell nanoseq workflow not to expect png plots. Let me know if you can think of a workaround. (and congrats on the little one!)

[ddomman]

I'm running into this same error while running our SARS-CoV-2 pipeline with Nanopore data using NFCORE_VIRALRECON

during NFCORE_VIRALRECON:NANOPORE:NANOPLOT

Caused by:
  Missing output file(s) `*.png` expected by process `NFCORE_VIRALRECON:NANOPORE:NANOPLOT (sample)`

Command executed:

  NanoPlot \
       \
      -t 2 \
      --fastq sample.fastq.gz
  cat <<-END_VERSIONS > versions.yml
  "NFCORE_VIRALRECON:NANOPORE:NANOPLOT":
      nanoplot: $(echo $(NanoPlot --version 2>&1) | sed 's/^.*NanoPlot //; s/ .*$//')
  END_VERSIONS

Command exit status:
  0

Command output:
  (empty)

@trum994 did you ever find a workaround for this?

Thanks!