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

[LDM #XZP-843172]: Fwd: LDM 6.13.14.26 is now on crunch01.allisonhouse.com, noaaport3.cod.edu

This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.

  • Subject: [LDM #XZP-843172]: Fwd: LDM 6.13.14.26 is now on crunch01.allisonhouse.com, noaaport3.cod.edu
  • Date: Sun, 11 Apr 2021 11:14:52 -0600
Hi Gilbert,

Did an "ldmadmin updategempaktables" and then stopped and then started LDM.
>
> That means our number of packets lost tomorrow will once again be larger
> than normal, but the good news is, so far, everything else is working fine.
> Will be interesting to see if this helps our packet loss, which is low
> anyway.
>
> And after further experimentation, it's doing this regardless of the size
> of the queue, unless it is the default.
>

What are you referring to by "it" and "this" in your last sentence?

So, knowing your system, and knowing mine, I matched your number just to
> check it out. Indeed, the machine paused for four or five seconds, shot the
> load up to five to six, but immediately came back down to one. A minute
> later, the load average was down to .64 and dropping. So, we'll see what
> happens. Unfortunately, because it is the start of a new Julian day, our
> data losses will be high because I made the LDM stop and start, and I also
> updated the GEMPAK tables back to current.
>
> And since I typed this out, the server load average is still dropping, now
> down to .40.
>
> Can someone explain why there is a substsntial load average surge for the
> first 10 or 15 seconds after starting the LDM with a NOAAport ingester, and
> then it quickly goes away? I would expect this from a relay server that is
> getting a large data feed such as level 2 radar or CONDUIT...but not
> NOAAport with no queue to backfill.
>

How did you change noaaportIngester(1)'s receive buffer size? Did you
change the system parameter "/proc/sys/net/core/rmem_default" multiple
times or did you change the system parameter "/proc/sys/net/core/rmem_max"
once and then change noaaportIngester(1)'s "-R" option argument multiple
times?

How many noaaportIngester(1) instances are running?

--Steve



Ticket Details
===================
Ticket ID: XZP-843172
Department: Support LDM
Priority: Normal
Status: Open
===================
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.