file_server_2 is wedged

[Mikael Pettersson]

On Wed, Sep 23, 2020 at 3:32 AM kuna.prime <kuna.prime@REDACTED> wrote:
>
> hi,
>
> is prim_file documented?

No.

If your problems occur during I/O as opposed to file-system operations
like stat, rename, or unlink, then opening your files with the "raw"
option might help.


>
>
> Sent with ProtonMail Secure Email.
>
> ‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐
> On Thursday, September 10, 2020 5:28 PM, Mikael Pettersson <mikpelinux@REDACTED> wrote:
>
> > On Thu, Sep 10, 2020 at 5:08 PM Matthew Evans mattevans123@REDACTED wrote:
> >
> > > Hi,
> > > We have a system running R21 where file_server_2 is stuck in a nif call. Here's the stack trace.
> > > root@REDACTED:~# doErlangFun "erlang:process_info(whereis(file_server_2),current_stacktrace)"
> > > {current_stacktrace,[{prim_file,read_info_1,3,[]},
> > > {file_server,handle_call,3,
> > > [{file,"file_server.erl"},{line,146}]},
> > > {gen_server,try_handle_call,4,
> > > [{file,"gen_server.erl"},{line,661}]},
> > > {gen_server,handle_msg,6,
> > > [{file,"gen_server.erl"},{line,690}]},
> > > {proc_lib,init_p_do_apply,3,
> > > [{file,"proc_lib.erl"},{line,249}]}]}
> > > The message queue of file_server_2 keeps growing.
> > > We have the async thread pool set to 25.
> > > Has anyone seen this before?
> >
> > Not this one specifically, but the file server has a very long history
> > of blocking processes. We often use prim_file instead to avoid that
> > problem. It may be better in newer OTP versions.
>
>