| Title: | DECmcc user notes file. Does not replace IPMT. |
| Notice: | Use IPMT for problems. Newsletter location in note 6187 |
| Moderator: | TAEC::BEROUD |
| Created: | Mon Aug 21 1989 |
| Last Modified: | Wed Jun 04 1997 |
| Last Successful Update: | Fri Jun 06 1997 |
| Number of topics: | 6497 |
| Total number of notes: | 27359 |
Hi,
I am having problems with exporting.
I started exporting some data on one of my node4s. I exported the
following:
Node4 FRED export period 12:00:00.00 (12 hrs)
Node4 FRED Line bna-0 export period 00:05:00.00 (5 mins)
Node4 FRED line ci-0 export period 00:05:00.00 (5 mins)
Node4 FRED line DMB-0 export period 00:05:00.00 (5 mins)
Node4 FRED circuit bna-0 export period 00:05:00.00 (5 mins)
Node4 FRED circui ci-0 export period 00:05:00.00 (5 mins)
Node4 FRED line DMB-0 export period 00:05:00.00 (5 mins)
After running this for a few days I found that there were
netserver.logs on FRED that totaled about 20,000 blocks.
The logs were being held open by nml server processes and in them
it had connect requests as often as every 5 secs -
Example: (PLKMCC is the MCC station)
--------------------------------------------------------
Connect request received at 4-NOV-1992 00:06:49.99
from remote process PLKMCC::"0=NETMGR"
for object "SYS$COMMON:[SYSEXE]NML.EXE"
--------------------------------------------------------
Connect request received at 4-NOV-1992 00:06:54.99
from remote process PLKMCC::"0=NETMGR"
for object "SYS$COMMON:[SYSEXE]NML.EXE"
--------------------------------------------------------
and on and on it went.
I have 2 questions :
1. Why is it connecting every 5 secs ??
2. Why are the nml server process holding the files open and also
creating new process and new files ??
I have had to stop exporting until I can solve this as among other
things the overhead it was creating on FRED was unacceptable.
Any help appreciated.
Jill
| T.R | Title | User | Personal Name | Date | Lines |
|---|---|---|---|---|---|
| 4034.1 | TOOK::SHMUYLOVICH | Thu Nov 05 1992 13:04 | 22 | ||
> 1. Why is it connecting every 5 secs ??
You have 6 exporting requests with 5 min period (I do not count
the first exporting requests with 12 hours period).
In order to get attribute data each exporting request makes
as many calls to the target entity as a number of partitions defined
for the target entity class. For Node4 Line and Node4 Circuit this
number is 7. Taking in consideration that each request for
Statistics attributes required another 4 calls we have the total
number of calls 10 (7 - 1 + 4). So one export operation requires 10
calls to the target entity. All 6 exporting requests make 60 calls
each 5 min or 1 call per 5 seconds.
Do you really need 5 min exporting period?
Sorry, I do not know the answer on your second question.
Sam
| |||||
| 4034.2 | Exporting... MAX LINKS exceeded... Links don't go away | KAOFS::BOIVIN | Moi, j'viens du nord! | Mon Nov 23 1992 14:07 | 49 |
Good day, I have one export command that is consuming all available links on my MCC station. Even when the exporting is completed, the links stick around until I kill the MCC process that did the historical exporting... Any ideas or is this QAR time? Thanks, Ed I had previously recorded all partitions for Node4 KAOR04 Line SYN-0. I now want to export: $ detmcc Welcome to DetMCC V1.3-14 � 1992 Digital Equipment AB DECmcc (V1.2.3) DetMCC> export node4 kaor04 line SYN-0 beg time = 15-nov-1992, - _DetMCC> end time = 11:00, - _DetMCC> in domain tro.s.mcc1.ct.mlokao Node4 57.42 Line SYN-0 AT 23-NOV-1992 13:44:36 Exporting completed DetMCC> Exit $ ncp sho known link Known Link Volatile Summary as of 23-NOV-1992 14:10:28 Link Node PID Process Remote link Remote user 16613 57.42 (KAOR04) 23E00E5E $DetMCC_EXEC 22924 NML 16614 57.42 (KAOR04) 23E00E5E $DetMCC_EXEC 22925 NML 8424 57.42 (KAOR04) 23E00E5E $DetMCC_EXEC 22927 NML 16617 57.42 (KAOR04) 23E00E5E $DetMCC_EXEC 22928 NML 16618 57.42 (KAOR04) 23E00E5E $DetMCC_EXEC 22929 NML 16620 57.42 (KAOR04) 23E00E5E $DetMCC_EXEC 22931 NML 16621 57.42 (KAOR04) 23E00E5E $DetMCC_EXEC 22932 NML .... *****************plus many more******************** | |||||