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

20040609: feeds off of (new)thelma (cont.)



>From:  Mike Schmidt <address@hidden>
>Organization:  UCAR/Unidata
>Keywords:  200406090722.i597Mr0A001713 IDD thelma power failure

Mike,

re: - number of upstream feed requests has increased: 17 -> 26 ?

>What's the deal with this?

I found that (new)thelma had a lot of feed requests to jackie that were
redundant to desi.  It also had a single request for CONDUIT from atm
and 5 requests to tgsv32 and two CRAFT requests that were doing nothing
since those OU machines had been decommissioned.

I dropped all redundant requests to jackie.

re: - remove the two CRAFT requests that are not viable anymore
     (remove requests to 129.15.194.23[12])

>Why aren't these viable?

OU decommissioned these machines.  They sent out a note to sites
feeding off of them to change their requests to drop these two.  We
changed thelma, but neglected to change newthelma (sigh).

re: - perhaps switch the CRAFT request to a single one from emo
     which is feeding from Purdue in a test mode

>Might as well

I have held off of this for the moment, but will keep it in mind
for future consideration.

re: - remove the redundant feed of CONDUIT data from tgsv32

>Couldn't hurt at this point

I dropped the reduntant ALTERNATE feeds of CONDUIT from tgsv32.

re: thoughts on dropping extra feeds

>I'd say go for it.

I decided to go for it before seeing your note.  The number of requests
dropped from 26 to 11/12 after I stopped and restarted the LDM.  The
problem is that the load average went up (!) instead of down.  It could
well be that we are in the transition where reconnecting processes are
working their way through the queue.  It could also be the case that
(new)thelma actually was able to ingest data, so there is more data to
relay to downstream sites.

>SCD has conserver up, but I'm locked out of the historical logging.
>I guess I'll have to wait until tomorrow to bring the 480 back online.

OK.  As long as (new)thelma and emo stay up, I think we will be OK.

By the way, the number of downstream feeds off of emo now stands at 53,
and the load average is low:

20040609.0800   1.13  1.97  2.56   53   9  62   1292 1478M 1024M
20040609.0801   2.45  2.18  2.60   53   9  62   1338 1476M 1024M
20040609.0802   5.00  2.96  2.87   53   9  62   1388 1475M 1024M
20040609.0803   2.70  2.69  2.77   53   9  62   1436 1474M 1024M
20040609.0804   1.59  2.38  2.65   53  10  63   1483 1472M 1024M
20040609.0805   2.06  2.39  2.63   53   9  62   1532 1471M 1024M
20040609.0806   2.29  2.48  2.65   53   9  62   1577 1512M 1024M
20040609.0807   2.05  2.39  2.61   53   9  62   1626 1505M 1024M
20040609.0808   1.33  2.14  2.50   53   9  62   1668 1502M 1024M
20040609.0809   1.03  1.85  2.36   53   9  62   1722 1500M 1024M
20040609.0810   0.48  1.52  2.19   53   9  62   1784 1497M 1024M

I am concerned that the load average on (new)thelma is not dropping (it
is now at 80 again).  It could be that there is a slug of CONDUIT data
coming in:

20040609.0755  53.47 32.46 42.15    0  11  11   6714 5125M 39M   93   1
20040609.0756  42.46 32.65 41.66   41  11  52   6753 4845M 184M    8   1
20040609.0757  54.83 37.88 42.91   67  11  78   6816 4713M 240M    9   1
20040609.0758  65.45 44.15 44.81   84  12  96   6881 4613M 271M    9   0
20040609.0759  70.94 49.29 46.60   83  12  95   6935 4555M 274M   11   0
20040609.0800  72.61 53.68 48.34   90  12 102   6998 4519M 284M   13   0
20040609.0801  76.95 58.31 50.34   92  12 104   7064 4474M 293M    8   0
20040609.0802  78.63 62.38 52.34   93  12 105   7124 4420M 294M   11   0
20040609.0803  78.84 65.18 53.94   94  12 106   7176 4387M 299M   11   0
20040609.0804  79.59 67.88 55.63   96  12 108   7239 4341M 299M   10   0
20040609.0805  78.92 69.80 57.11   95  12 107   7297 4307M 299M   14   0
20040609.0806  80.96 71.98 58.72   96  11 107   7356 4273M 301M   11   0
20040609.0807  80.44 73.48 60.09   94  12 106   7417 4225M 299M    6   0
20040609.0808  78.17 74.14 61.22   95  12 107   7476 4180M 302M    7   0
20040609.0809  78.11 74.86 62.33   95  12 107   7541 4136M 302M    7   0
20040609.0810  78.06 75.44 63.35   95  12 107   7603 4076M 302M    6   0
20040609.0811  80.25 76.43 64.42   95  12 107   7654 4031M 303M    6   0

The funny thing about working on this late is that I have been
answering emails from EUMETSAT.  Given the time difference, I am
chatting with someone who is at work in her morning :-)

Tom
--
+-----------------------------------------------------------------------------+
* Tom Yoksas                                             UCAR Unidata Program *
* (303) 497-8642 (last resort)                                  P.O. Box 3000 *
* address@hidden                                   Boulder, CO 80307 *
* Unidata WWW Service                             http://www.unidata.ucar.edu/*
+-----------------------------------------------------------------------------+


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.