Discussion:
[OTAS-Planning]quakerpt news
Brian Butcher
2005-11-02 16:09:46 UTC
Permalink
I have just make a new release of the quakerpt software with changes designed
to make it a tiny bit easier for Ben and Jose to get it working.

You should see reference to more than one peer within 5 minutes of startup of
quakereport.py - I will try to keep at least two peers running for the next
few week (The peers have the same IP No but different port no). Hopefully no
editing of the config.py file will be now be needed for a client on its own.

http://prdownloads.sourceforge.net/etad/quakerpt-alpha-0.2.tar.gz?download

As we may want to get discussion on this away off the otas-planning list I
have created a new list on etad-***@lists.sourceforge.net - It will
become possible to subscribe in 6-12 hours from now.

Brian
Ben Middleton
2005-11-02 18:33:46 UTC
Permalink
looks good, I see reference to several hosts:

$ python quakereport.py
starting quakerpt.py version: 0.1
running on 82.147.17.23:8085
82.147.17.23:8080:Etad's parsing of USGS RSS feed
checking time for peer 82.147.17.23:8080
peer time is 2005-11-02T18:19:38GMT
mytime is 2005-11-02T18:19:38GMT
failed to authenticate myself at 82.147.17.23:8080 with c1353c1
new peer 82.147.17.23:8084
Adding host 82.147.17.23:8084 to database
setting peer.conntime to 1970-01-01T00:00:00GMT
new peer 82.147.17.23:8081
Adding host 82.147.17.23:8081 to database
Post by Brian Butcher
I have just make a new release of the quakerpt software with changes designed
to make it a tiny bit easier for Ben and Jose to get it working.
You should see reference to more than one peer within 5 minutes of startup of
quakereport.py - I will try to keep at least two peers running for the next
few week (The peers have the same IP No but different port no). Hopefully no
editing of the config.py file will be now be needed for a client on its own.
http://prdownloads.sourceforge.net/etad/quakerpt-alpha-0.2.tar.gz?
download
As we may want to get discussion on this away off the otas-planning list I
become possible to subscribe in 6-12 hours from now.
Brian
-------------------------------------------------------
Tame your development challenges with Apache's Geronimo App Server.
Download
it for free - -and be entered to win a 42" plasma tv or your very own
Sony(tm)PSP. Click here to play: http://sourceforge.net/geronimo.php
_______________________________________________
OTAS-Planning mailing list
https://lists.sourceforge.net/lists/listinfo/otas-planning
Ben Middleton
2005-11-02 18:48:53 UTC
Permalink
oops, just noticed the link (I really should read my email more
carefully)
the results below are using the OLD quakereport code, not the new
code, which I haven't downloaded yet. Sorry about any confusion caused
by this.

I also noticed that memory usage seems rather larger than advertised
(although perhaps I shouldn't criticise, given my efforts in this
arena)

relevant output from top:

PID COMMAND %CPU TIME #TH #PRTS #MREGS RPRVT RSHRD RSIZE
VSIZE
9250 python 0.0% 0:17.11 1 13 146 2.10M 1.33M 3.28M
30.2M


this python process comes from:

$ python quakereport.py

ie:
$ ps -auxwww | grep 9250
benmiddl 9250 0.0 0.6 30912 3360 p6 S+ 6:19PM 0:17.13
python quakereport.py

I'm not sure if this is a peculiarity of mac os x, or whether I'm
misinterpreting the output of top or not.
Post by Ben Middleton
$ python quakereport.py
starting quakerpt.py version: 0.1
running on 82.147.17.23:8085
82.147.17.23:8080:Etad's parsing of USGS RSS feed
checking time for peer 82.147.17.23:8080
peer time is 2005-11-02T18:19:38GMT
mytime is 2005-11-02T18:19:38GMT
failed to authenticate myself at 82.147.17.23:8080 with c1353c1
new peer 82.147.17.23:8084
Adding host 82.147.17.23:8084 to database
setting peer.conntime to 1970-01-01T00:00:00GMT
new peer 82.147.17.23:8081
Adding host 82.147.17.23:8081 to database
Post by Brian Butcher
I have just make a new release of the quakerpt software with changes designed
to make it a tiny bit easier for Ben and Jose to get it working.
You should see reference to more than one peer within 5 minutes of startup of
quakereport.py - I will try to keep at least two peers running for the next
few week (The peers have the same IP No but different port no). Hopefully no
editing of the config.py file will be now be needed for a client on its own.
http://prdownloads.sourceforge.net/etad/quakerpt-alpha-0.2.tar.gz?
download
As we may want to get discussion on this away off the otas-planning list I
become possible to subscribe in 6-12 hours from now.
Brian
-------------------------------------------------------
Tame your development challenges with Apache's Geronimo App Server.
Download
it for free - -and be entered to win a 42" plasma tv or your very own
Sony(tm)PSP. Click here to play: http://sourceforge.net/geronimo.php
_______________________________________________
OTAS-Planning mailing list
https://lists.sourceforge.net/lists/listinfo/otas-planning
Brian Butcher
2005-11-02 20:48:08 UTC
Permalink
Post by Ben Middleton
Post by Ben Middleton
$ python quakereport.py
starting quakerpt.py version: 0.1
running on 82.147.17.23:8085
82.147.17.23:8080:Etad's parsing of USGS RSS feed
checking time for peer 82.147.17.23:8080
peer time is 2005-11-02T18:19:38GMT
mytime is 2005-11-02T18:19:38GMT
failed to authenticate myself at 82.147.17.23:8080 with c1353c1
new peer 82.147.17.23:8084
Adding host 82.147.17.23:8084 to database
setting peer.conntime to 1970-01-01T00:00:00GMT
new peer 82.147.17.23:8081
Adding host 82.147.17.23:8081 to database
Great! - thanks - I can see you on all public peers now
Post by Ben Middleton
I also noticed that memory usage seems rather larger than advertised
(although perhaps I shouldn't criticise, given my efforts in this
arena)
PID COMMAND %CPU TIME #TH #PRTS #MREGS RPRVT RSHRD RSIZE
VSIZE
9250 python 0.0% 0:17.11 1 13 146 2.10M 1.33M 3.28M
30.2M
$ python quakereport.py
$ ps -auxwww | grep 9250
benmiddl 9250 0.0 0.6 30912 3360 p6 S+ 6:19PM 0:17.13
python quakereport.py
I'm not sure if this is a peculiarity of mac os x, or whether I'm
misinterpreting the output of top or not.
Interesting: for me this python usage is too small to register on top. I can
only see it on ps aux. On etad itself, running 3 peers plus the usual etad
plus an apache server, email server and mysql server python usage is trivial.
i.e.

top - 20:38:48 up 5 days, 7:01, 8 users, load average: 0.02, 0.09, 0.08
Tasks: 70 total, 1 running, 69 sleeping, 0 stopped, 0 zombie
Cpu(s): 0.3% user, 0.3% system, 0.0% nice, 99.3% idle
Mem: 224464k total, 133172k used, 91292k free, 1456k buffers
Swap: 730948k total, 17776k used, 713172k free, 105272k cached

I can only think that this must be a consequence of different operating
systems. Following is from my desktop machine with two peers running:

server gives similar statistics.

***@bonnynuts:~$ ps aux | grep python
tester 2212 0.0 0.4 6660 4352 pts/4 S+ 12:57 0:00 python
server.py
tester 2227 0.0 0.4 6660 4352 pts/4 S+ 12:59 0:00 python
server.py
tester 2245 0.0 0.4 6672 4268 pts/5 S+ 13:04 0:13 python
quakereport.py
brian 3455 0.0 0.4 6636 4208 pts/6 S+ 18:19 0:04 python
quakereport.py
brian 3728 0.0 0.0 1820 576 pts/2 R+ 20:17 0:00 grep python

- Total declared memory usage by python less than 30 K on either machine!

Odd

My first though is that there may be a difference in shared library usage
between Mac and Linux. Can your OS handle shared C libraries or does it have
to link statically when using Python .pyc files?

Certainly my Linux experience has not been good when I've used static linking.

Brian

Is it possible that the effect has occured the other way when running Java on
Linux?
Ben Middleton
2005-11-02 21:29:35 UTC
Permalink
I've cc'd onto etad-developer as this discussion is probably more
relevant to that list
Post by Brian Butcher
Post by Ben Middleton
Post by Ben Middleton
$ python quakereport.py
starting quakerpt.py version: 0.1
running on 82.147.17.23:8085
82.147.17.23:8080:Etad's parsing of USGS RSS feed
checking time for peer 82.147.17.23:8080
peer time is 2005-11-02T18:19:38GMT
mytime is 2005-11-02T18:19:38GMT
failed to authenticate myself at 82.147.17.23:8080 with c1353c1
new peer 82.147.17.23:8084
Adding host 82.147.17.23:8084 to database
setting peer.conntime to 1970-01-01T00:00:00GMT
new peer 82.147.17.23:8081
Adding host 82.147.17.23:8081 to database
Great! - thanks - I can see you on all public peers now
Post by Ben Middleton
I also noticed that memory usage seems rather larger than advertised
(although perhaps I shouldn't criticise, given my efforts in this
arena)
PID COMMAND %CPU TIME #TH #PRTS #MREGS RPRVT RSHRD RSIZE
VSIZE
9250 python 0.0% 0:17.11 1 13 146 2.10M 1.33M 3.28M
30.2M
$ python quakereport.py
$ ps -auxwww | grep 9250
benmiddl 9250 0.0 0.6 30912 3360 p6 S+ 6:19PM 0:17.13
python quakereport.py
I'm not sure if this is a peculiarity of mac os x, or whether I'm
misinterpreting the output of top or not.
Interesting: for me this python usage is too small to register on top. I can
only see it on ps aux. On etad itself, running 3 peers plus the usual etad
plus an apache server, email server and mysql server python usage is trivial.
i.e.
top - 20:38:48 up 5 days, 7:01, 8 users, load average: 0.02, 0.09, 0.08
Tasks: 70 total, 1 running, 69 sleeping, 0 stopped, 0 zombie
Cpu(s): 0.3% user, 0.3% system, 0.0% nice, 99.3% idle
Mem: 224464k total, 133172k used, 91292k free, 1456k buffers
Swap: 730948k total, 17776k used, 713172k free, 105272k cached
I can only think that this must be a consequence of different operating
server gives similar statistics.
tester 2212 0.0 0.4 6660 4352 pts/4 S+ 12:57 0:00 python
server.py
tester 2227 0.0 0.4 6660 4352 pts/4 S+ 12:59 0:00 python
server.py
tester 2245 0.0 0.4 6672 4268 pts/5 S+ 13:04 0:13 python
quakereport.py
brian 3455 0.0 0.4 6636 4208 pts/6 S+ 18:19 0:04 python
quakereport.py
brian 3728 0.0 0.0 1820 576 pts/2 R+ 20:17 0:00 grep python
- Total declared memory usage by python less than 30 K on either machine!
Odd
My first though is that there may be a difference in shared library usage
between Mac and Linux. Can your OS handle shared C libraries or does it have
to link statically when using Python .pyc files?
I don't know but I'll do some research. Sounds plausible though.
Post by Brian Butcher
Certainly my Linux experience has not been good when I've used static linking.
Brian
Is it possible that the effect has occured the other way when running Java on
Linux?
it's possible, though I'm pretty sure the memory usage was fairly
significant on my machine too. I haven't heard of any such problems
occurring, however. Again, I'll have to do some research to answer
that one.
Post by Brian Butcher
-------------------------------------------------------
Tame your development challenges with Apache's Geronimo App Server.
Download
it for free - -and be entered to win a 42" plasma tv or your very own
Sony(tm)PSP. Click here to play: http://sourceforge.net/geronimo.php
_______________________________________________
OTAS-Planning mailing list
https://lists.sourceforge.net/lists/listinfo/otas-planning
Loading...