[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
20050225:Latency issues UNI
- Subject: 20050225:Latency issues UNI
- Date: Fri, 25 Feb 2005 12:26:33 -0700 (MST)
Hi Patrick,
Not so pretty...looking at OU, your feed site, I see much latency.
IDS
http://my.unidata.ucar.edu/cgi-bin/rtstats/iddstats_nc?IDS|DDPLUS+stokes.metr.ou.edu
NNEXRAD
http://my.unidata.ucar.edu/cgi-bin/rtstats/iddstats_nc?NNEXRAD+stokes.metr.ou.edu
If we look at our machines:
IDS
http://my.unidata.ucar.edu/cgi-bin/rtstats/iddstats_nc?IDS|DDPLUS+oliver.unidata.ucar.edu
NNEXRAD
http://my.unidata.ucar.edu/cgi-bin/rtstats/iddstats_nc?NNEXRAD+oliver.unidata.ucar.edu
We do not see it, I'll contact OU and see what is going on...
Who is your failover...? You may want to go there until we see what is up
with OU.
If you do not have a failover, and these latencies are causing problems,m
please feel free to feed from:
idd.unidata.ucar.edu
in the meantime.
caveat...We are doing a power down on the Mesa this weekend, so things may
get flaky here ;(..especially from 6-10 pm MST..this ~should not affect
idd.unidata.ucar.edu
as it lives here, but.....
Keep me posted,
Jeff
---------------------------------------------------------------------
Jeff Weber address@hidden :
Unidata Program Center PH:303-497-8676 :
University Corp for Atmospheric Research 3300 Mitchell Ln :
http://www.unidata.ucar.edu/staff/jweber Boulder,Co 80307-3000 :
---------------------------------------------------------------------
On Fri, 25 Feb 2005, Unidata Support wrote:
>
> ------- Forwarded Message
>
> >To: address@hidden
> >From: "Patrick O'Reilly" <address@hidden>
> >Subject: Latency Issues
> >Organization: UCAR/Unidata
> >Keywords: 200502251912.j1PJC6v2021287
>
> Hello,
>
> I have been having intermittent latency issues since the 23rd. I've made no
> changes to my ldm setup. At times it's all feedtypes, at other times it's
> the HDS feed that's behind, yet others, it's the NNEXRAD feed that's
> behind...etc. My traceroute and ping output look reasonable, no lost
> packets, no long delays to my feed machine. Top output shows nothing out of
> the ordinary, 4 cpus running 95-100% idle most times. I ran a backup
> machine to see if the problem was with my main ldm machine, but no, they're
> both backed up. From my main machine:
>
> Feb 25 18:44:53 pqutil: 110 20050225174639.600 IDS|DDPLUS 24114560
> SAUS45 KCYS 251746 /pMTRBRX
> Feb 25 18:44:54 pqutil: 2604 20050225184452.817 WMO 518 SDUS81 KBTV
> 251835 /pDPACXX
> Feb 25 18:44:54 pqutil: 148 20050225184452.926 IDS|DDPLUS 508 NXUS68
> PAFC 251843 /pGSMAHG
> Feb 25 18:44:54 pqutil: 19325 20050225184448.677 HDS 421 YTWE55 KWBG
> 251800 /mRUC2
> Feb 25 18:44:54 pqutil: 5464 20050225184453.050 WMO 519 SDUS84 KOUN
> 251842 /pDPAFDR
>
> and sometimes even a single feed has much different latencies:
>
> Feb 25 19:03:28 pqutil: 292 20050225190320.749 IDS|DDPLUS 383 SXUS70
> KWAL 251901
> Feb 25 19:03:28 pqutil: 76 20050225182035.682 IDS|DDPLUS 5608166
> SRCA20 KWAL 251819
> Feb 25 19:03:28 pqutil: 75 20050225182035.683 IDS|DDPLUS 5608167
> SRTX20 KWAL 251819
> Feb 25 19:03:28 pqutil: 128 20050225182035.684 IDS|DDPLUS 5608168
> SRGA20 KWAL 251819
> Feb 25 19:03:28 pqutil: 75 20050225182035.684 IDS|DDPLUS 5608169
> SXHN40 KWAL 251819
>
> and from the backup machine:
>
> Feb 25 18:42:41 pqutil: 399 20050225174124.871 IDS|DDPLUS 24111253
> SXUS28 KWOH 251739 /pRRSSGF
> Feb 25 18:42:41 pqutil: 2253 20050225184240.689 NNEXRAD 767 SDUS33 KOAX
> 251832 /pNVWOAX
> Feb 25 18:42:41 pqutil: 330389 20050225184151.017 WMO 017 TIGF02 KNES
> 251830
> Feb 25 18:42:41 pqutil: 7931 20050225184241.007 HDS 688 ZPQB94 KWBG
> 251800 /mRUC
>
> so now, it's the IDS|DDPLUS feed that's behind. My queues are big enough on
> both machines for an hours worth of data. See my issues with the main
> machine:
>
> http://my.unidata.ucar.edu/cgi-bin/rtstats/iddstats_nc?IDS|DDPLUS+thunder.st
> orm.uni.edu
>
> and the backup:
>
> http://my.unidata.ucar.edu/cgi-bin/rtstats/iddstats_nc?IDS|DDPLUS+blizzard.s
> torm.uni.edu
>
> In my ldmd, there's lots of:
>
> Feb 25 18:48:05 thunder pqact[2672]: pbuf_flush 33: time elapsed 2.275563
> Feb 25 18:48:43 thunder pqact[2672]: child 3979 exited with status 127
> Feb 25 18:48:55 thunder pqact[2672]: child 3982 exited with status 127
> Feb 25 18:49:01 thunder pqact[2672]: child 3984 exited with status 127
> Feb 25 18:49:04 thunder pqact[2672]: child 3990 exited with status 127
> Feb 25 18:50:25 thunder pqact[2672]: pbuf_flush 35: time elapsed 4.091240
> Feb 25 18:50:39 thunder pqact[2672]: child 3994 exited with status 127
> Feb 25 18:50:55 thunder pqact[2672]: child 3996 exited with status 127
> Feb 25 18:52:43 thunder pqact[2672]: pbuf_flush 35: time elapsed 3.555103
>
> and in dcgrib.log, lots of:
>
> [3273] 050225/1254[DCGRIB 1] Grid navigation 255 incompatible with file
> data/gempak/model/awc/20050225_icing.gem
> [3273] 050225/1254[DCGRIB 1] Grid navigation 255 incompatible with file
> data/gempak/model/awc/20050225_icing.gem
> [3273] 050225/1254[DCGRIB 1] Grid navigation 255 incompatible with file
> data/gempak/model/awc/20050225_icing.gem
>
> with different files in the spot where it says 20050225_icing.gem.
>
> The main machine (thunder) is on a super fast connection and it's a
> dual-Xeon 3.0 GHz machine with 2 GB memory and 240 Gb storage. The machine
> has always been fine until about Wednesday. I thought I would check with
> you all before I start bothering the university network guys. Any ideas
> what to look at? Or have there been changes anywhere since Wednesday that I
> missed?
>
> Patrick
>
> --
> 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.
>
> ------- End of Forwarded Message
>