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

20030327: rtstats

Chuck,

Thanks for getting rtstats up and running. 
In this beta release of LDM6, rtstats is still using LDM 5 protocol to send
the statistics back. We are currently reworking parts of ldmsend to
allow it to use the LDM 6 protocol which will make for major performance
improvements, and we should have direct opening to port 388
(and only look for portmapper if that fails).  

The stats page is currently under reconstruction to a central database now
that we are approaching enough sites sending us stats in order to 
do more topology information, so not all feeds are being listed
(in order to conserve resources during testing).


Steve Chiswell





>From: address@hidden
>Organization: UCAR/Unidata
>Keywords: 200303271620.h2RGKTB2022824

>
>We are running ldm 6.0.2 on a Sun Solaris 5.8 box.  Here are a couple
>of points you probably already know...I'm just passing them along:
>
>1. On the terminal connected to the ldm processes we would find the following
>line:
>
>could not sign on 130
>
>I traced this back to a "printf" command in ldmsend.c line 446.  I
>simply commented this line out...nothing major just a nuisance.
>
>2. We find the following message numerous times in the ldmd.log
>
>Mar 27 10:22:19 inflow.eas.slu.edu rtstats[1274]: clnt_create(rtstats.unidata.
> ucar.edu, 300029, 5, "tcp"): rtstats.unidata.ucar.edu: RPC: Miscellaneous tli
>  error - An event requires attentionInterrupted system call
>
>3. It appears now the rtstats web page is only tracking (for us):
>HDS, IDS|DDPLUS, NNEXRAD, UNIWISC  (we do not request CONDUIT data)
>
>
>On a potential more important point.  Our systems guru had to open up
>port 111 for rtstats to work.  He was not thrilled by this, apparently
>this can be a security hole in Suns.  He has already noticed other "activity"
>with this port.  He has informed me that port 111 will not remain open
>indefinitely, he would like to shut it down in the near future....
>Is there a work-around or are we heading for trouble?
>
>Chuck
>