[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[LDM #FLA-146766]: ldm file writes



Steve,

> Ah, so the fact that there was only one fl_removeAndFree() indicated that
> the -flush was still active until I did the stop/clean/start. We have 5
> pqact processes with individual .conf files. The metar stuff is in
> the pqact.conf file, which I assumed is the one that is acted on by
> a ldmadmin pqactHUP. I guess I could be wrong.

Actually, an "ldmadmin pqactHUP" should cause every pqact(1) process to
receive a SIGHUP.

> I can now confirm that I'm seeing fl_removeAndFree() for both local disk
> and network metar data writes. Local disk files continue to write normally
> now. So... I'm cautiously optimistic we can close this ticket. The problem
> seems to have been with the -flush STDIOFILE option. Why that suddenly
> became a problem a few days ago is beyond me.

The operating system determines what a "flush" means. Did you upgrade yours
recently?

> One last thing to tidy up this thread. I reviewed our pqact*.conf files.
> Those local disk metar writes were the only instance of using the -flush
> option with the STDIOFILE action. All other instances of STDIOFILE used
> -close. We also use -flush with many instances of the FILE action. No
> problems there. So... could be a glitch with STDIOFILE and -flush.

I don't see how. I just examined the code and the processing of the "-flush"
option is, effectively, the same for both the FILE and STDIOFILE actions.

It's a mystery.

Regards,
Steve Emmerson

Ticket Details
===================
Ticket ID: FLA-146766
Department: Support LDM
Priority: Normal
Status: Closed
===================
NOTE: All email exchanges with Unidata User Support are recorded in the Unidata 
inquiry tracking system and then made publicly available through the web.  If 
you do not want to have your interactions made available in this way, you must 
let us know in each email you send to us.