Menu
▾
▴
perfparse-devel-int — Developers Internal high traffic discussion PP development. Contributers only.
You can subscribe to this list here.
| 2004 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(19) |
Dec
(72) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2005 |
Jan
(35) |
Feb
(36) |
Mar
|
Apr
|
May
(3) |
Jun
(2) |
Jul
|
Aug
(4) |
Sep
|
Oct
(1) |
Nov
|
Dec
|
| 2006 |
Jan
(2) |
Feb
|
Mar
(2) |
Apr
(11) |
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(1) |
Dec
|
|
From: Yves <yme...@pe...> - 2004-11-30 09:37:12
|
> The "perfparse-0.104.1ym1" is > Not found on > http://pagesperso.laposte.net/ymettier/perfparse-devel You found it thanks to my other email :) For the doc, check the last version here : http://pagesperso.laposte.net/ymettier/perfparse-devel/perfparse-doc/ ("Install guide in one html page per section" in not available yet) Ben, this will be in 0.104.1ym2 for future 0.104.2 :) Comments below... > I did check out the perfparse-doc while I was there. > Here is some suggestions to hopefully > To help improve the documentation for: > > Chapter5: Method 4. > > To create a perfparse.cfg config file using the defaults: > ./perfparsed --show_config # To see current config. > Or to create a new 1 to compare to current: > ./perfparsed --show_config > /var/tmp/perfparse.cfg Those lines were improved and moved to the chapter <using perfparse> :) > With nagios-1.2 edit misccommands.cfg and comment out the following > Definitions for host and service performance data: > Define command{ > command_name process-host-perfdata > .... > } > and > > Define command{ > Command_name process-service-perfdata > .... > } > > > Then in the nagios.cfg verify the following are in the file: > > cfg_file=3D/usr/local/nagios/etc/nagios_perfparse.cfg > process_performance_data=3D1 > host_perfdata_command=3Dprocess-host-perfdata > service_perfdata_command=3Dprocess-service-perfdata > > In the perfparse.cfg make sure the the variable Service_log is set to: > Service_log =3D "|/path/to/the/pipe" > Example: > Service_log =3D > "|/usr/local/nagios/var/rw/serviceperf.log > Note: > Permissions for "serviceperf.log" need to be set so nagios daemon can > Write to the FIFO. Also make sure serviceperf.log file is not present > When starting the perfparsed. the fifo can already exist with 0.104.1. You are true with 0.103.X :) No more comments below. Thanks for your contribution, Yves > Finally in the nagios_perfparse.cfg file: > > define_command{ > command_name process-service-perfdata > command_line $USER2$/bin/perfparse_nagios_pipe.command.pl > /path/to/the/pipe "$TIMET$" "$HOSTNAME$" "$SERVICEDESC$" "$OUTPUT$" > "$SERVICESTATE$" "$PERFDATA$" > } > > Example: > > Define_command{ > command_name process-service-perfdata > command_line $USER2$/bin/perfparse_nagios_pipe.command.pl > /usr/local/nagios/var/rw/serviceperf.log "$TIMET$" "$HOSTNAME$" > "$SERVICEDESC$" "$OUTPUT$" "$SERVICESTATE$" "$PERFDATA$" > } > > > > > -----Original Message----- > From: Yves [mailto:yme...@pe...] > Sent: Monday, November 29, 2004 2:16 PM > To: Stu...@am... > Subject: RE: [Perfparse-users] perfparsed & CPU usage > > Tim found the fix for that bug. > You can get perfparse-0.104.1ym1 on > http://pagesperso.laposte.net/ymettier/perfparse-devel/ with the fix, a= nd > perfparse-0.104.2 will probably be released soon. > > Thanks for the feedback ! > Yves > > > >> I also have this behavior on an Ultra1 running >> Solaris 8 2/02. Perfparse version .0.103.2 and >> Nagios 1.2 >> >> I made the select() &tm change to NULL; >> But I didn't see a positive change. >> I still see the perfparsed loads of 95%+ % >> >> I got a truss for you >> >> As root I issued the following command >> truss -o /var/tmp/perparse0.103.2.truss -f ./perfparsed -d >> >> About a 835 lines get written to the file. >> >> I then changed the permissions to 666 on >> nagios/var/rw/serviceperf.log # fifo >> >> and started nagios. >> >> I would try out 104.1 but I'm getting >> a compile error for libpp_storage_mysql.so.0 >> >> "Text relocation remains referenced >> against symbol offset in file >> <unknown> 0x884 >> /usr/local/mysql/lib/libmysqlclient.a(client.o)" >> >> >> -----Original Message----- >> From: Yves Mettier [mailto:yme...@li...] >> Sent: Friday, November 26, 2004 3:35 AM >> To: per...@li... >> Subject: Re: [Perfparse-users] perfparsed & CPU usage >> >> >>> Hello, >>> I configured perfparsed to read from a pipe. (Method 4) (config below= ) >>> I noticed that the CPU usage for perfparse is constantly at > 95% (ss= e >>> part of 'top' output below). >>> >>> Without perfparsed, top reports 99.7% idle, with perfparsed, it >>> reports about 50% idle. I should add that this is a test setup, with >>> nagios checking only 60 services every 5 minutes (iow, perfparsed >>> takes up a lot of CPU for doing close to nothing) >> >> In perfparse/log_reader.c, we use select(). There is a timeout of one >> second. >> On the line with select(), the last argument is "&tm". Could you repla= ce > it >> with "NULL" >> and test ? There should be no side effect with this change. >> >>> the production setup checks around 600 services, so I'm a little worr= ied. >>> >>> Any thoughts, insights ? >> >> If it is not a bad use of select(), I have no idea. Give us more, with= for >> example using >> strace (linux) or truss (solaris). >> >> >> Yves >> >> -- >> - Homepage - http://ymettier.free.fr - http://www.logicacmg.com - >> - GPG key - http://ymettier.free.fr/gpg.txt - >> - Maitretarot - http://www.nongnu.org/maitretarot/ - >> - Perfparse - http://perfparse.sf.net/ - >> >> >> >> ------------------------------------------------------- >> SF email is sponsored by - The IT Product Guide >> Read honest & candid reviews on hundreds of IT Products from real user= s. >> Discover which products truly live up to the hype. Start reading now. >> http://productguide.itmanagersjournal.com/ >> _______________________________________________ >> Perfparse-users mailing list >> Per...@li... >> https://lists.sourceforge.net/lists/listinfo/perfparse-users >> >> > > > -- > - Homepage - http://ymettier.free.fr - http://www.logicacmg.com - > - GPG key - http://ymettier.free.fr/gpg.txt - > - Maitretarot - http://www.nongnu.org/maitretarot/ - > - Perfparse - http://perfparse.sf.net/ - > > --=20 - Homepage - http://ymettier.free.fr - http://www.logicacmg.com - - GPG key - http://ymettier.free.fr/gpg.txt - - Maitretarot - http://www.nongnu.org/maitretarot/ - - Perfparse - http://perfparse.sf.net/ - |
|
From: Flo G. <fl...@bi...> - 2004-11-29 22:16:55
|
Had a relaying problem right now so the mail was rejected - have to switch private mail to my own mailserver :-) So, i'm back now from basketball - how about a tiki tomorrow? Flo On Mon, 29 Nov 2004, Flo Gleixner wrote: > > > hey, just almost missed the wiki discussion :-) > > Since i'm a project admin of tikiwiki, i'd surely vote for tikiwiki. I also > have a server with root access and 500GB traffic fast ethernet connection > (100MBit), own dns server, mailserver, ... whatever one can do with a root > account. I'd be pleased to host a perfparse site (or part of it). I can > install can configure a tikiwiki in less than 10 mins :-) But at the momenet > i have to go to play basketball. I'll be back in 3-4 hours... > > Flo > > > On Mon, 29 Nov 2004, Yves wrote: > >> >>> What structure would you like to see in a wiki/tiki system? >>> >>> For my self I would like to see editable specifications of new work we >>> are going to undertake. Where comments as well as actual changes can be >>> made. >> >> Specifications, when there are some. Yes (I'm thinking about the next >> database schema >> for mysql, and API for modules that I will define soon). >> But don't take them as the reference. The reference is always the code : >> - create_mysql.sh for mysql database schema (Ben, do you have better idea >> than that file ?) >> - storage.h for storage modules. >> >> But I like the idea to have a public draft for specifications. >> >> >>> Yves also mentioned a public FAQ, private edit. Now the software is >>> gaining in ability, this sounds like an excellent idea. >> >> have a look on that : http://www.squirrelmail.org/wiki/UserFAQ >> >> Answering the question at the beginning of this mail, I have 2 suggestions >> : >> - have one wiki with entries for the main stuff (spec drafts, faq, ...) >> - have one wiki per main stuff (one for spec drafts, one for faq, ...) >> >> I suggest one wiki because we can link some FAQ entries inside our drafts >> ? >> >> btw, if you don't know what wiki to use, have a look on the wiki that >> squirrelmail uses, >> and have a look on what wikipedia uses. I don't know which is the best. >> >>> Thanks to Tim and Yves for finding and fixing a bug so quickly. >> >> Thanks to Tim : he noticed the problem, found the bug, well, yes, that's >> me who found >> the bug, and he found the solution while I was actively searching for it >> :) >> >> Yves >> >> -- >> - Homepage - http://ymettier.free.fr - http://www.logicacmg.com - >> - GPG key - http://ymettier.free.fr/gpg.txt - >> - Maitretarot - http://www.nongnu.org/maitretarot/ - >> - Perfparse - http://perfparse.sf.net/ - >> >> >> >> ------------------------------------------------------- >> SF email is sponsored by - The IT Product Guide >> Read honest & candid reviews on hundreds of IT Products from real users. >> Discover which products truly live up to the hype. Start reading now. >> http://productguide.itmanagersjournal.com/ >> _______________________________________________ >> Perfparse-devel-int mailing list >> Per...@li... >> https://lists.sourceforge.net/lists/listinfo/perfparse-devel-int >> > |
|
From: Yves <yme...@pe...> - 2004-11-29 17:10:15
|
> What structure would you like to see in a wiki/tiki system? > > For my self I would like to see editable specifications of new work we > are going to undertake. Where comments as well as actual changes can b= e > made. Specifications, when there are some. Yes (I'm thinking about the next dat= abase schema for mysql, and API for modules that I will define soon). But don't take them as the reference. The reference is always the code : - create_mysql.sh for mysql database schema (Ben, do you have better idea= than that file ?) - storage.h for storage modules. But I like the idea to have a public draft for specifications. > Yves also mentioned a public FAQ, private edit. Now the software is > gaining in ability, this sounds like an excellent idea. have a look on that : http://www.squirrelmail.org/wiki/UserFAQ Answering the question at the beginning of this mail, I have 2 suggestion= s : - have one wiki with entries for the main stuff (spec drafts, faq, ...) - have one wiki per main stuff (one for spec drafts, one for faq, ...) I suggest one wiki because we can link some FAQ entries inside our drafts= ? btw, if you don't know what wiki to use, have a look on the wiki that squ= irrelmail uses, and have a look on what wikipedia uses. I don't know which is the best. > Thanks to Tim and Yves for finding and fixing a bug so quickly. Thanks to Tim : he noticed the problem, found the bug, well, yes, that's = me who found the bug, and he found the solution while I was actively searching for it = :) Yves --=20 - Homepage - http://ymettier.free.fr - http://www.logicacmg.com - - GPG key - http://ymettier.free.fr/gpg.txt - - Maitretarot - http://www.nongnu.org/maitretarot/ - - Perfparse - http://perfparse.sf.net/ - |
|
From: Ben C. <BCl...@pe...> - 2004-11-29 16:28:45
|
Cook, Garry wrote: > I've not used a WIKI/TIKI, but I hear that they are not too difficult to > setup. Think it requires a backend DB, although not positive. Like you > mentioned below, sf.net will give access to MySQL, but I believe we need > to make a formal request for this. I will see what database we need, and what services sf.net offer. Tiki does use MySQL which I know sf offer. I just have to work out how :) I've not installed tiki before, so I'm in the dark as well. Some fun for the weekend maybe :) To all developers: What structure would you like to see in a wiki/tiki system? For my self I would like to see editable specifications of new work we are going to undertake. Where comments as well as actual changes can be made. Yves also mentioned a public FAQ, private edit. Now the software is gaining in ability, this sounds like an excellent idea. Thanks to Tim and Yves for finding and fixing a bug so quickly. Regards, Ben > > Garry W. Cook, CCNA > Network Infrastructure Manager > MACTEC, Inc. - http://www.mactec.com/ > 303.273.5050 (Office) - 720.220.1862 (Mobile) > > Ben Clewett wrote: > >>Garry, >> >>One thing that Yves talked about was a tiki system. Have you ever set >>one up? This could be really useful for, as Yves suggested, an FAQ >>and a private document of agreed bits. sf.net give access to the >>database, but not looked any further at it... >> >>Ben >> >>Cook, Garry wrote: >> >> >>>Sorry, I haven't been following this thread very closely, I'll have >>>to go back and read it all the way through. However, to add a page, >>>this is not too hard. I'm sure Flo can do it, as he has been working >>>with PHP a bit, the pages are written in PHP. Ben has been >>>maintaining them in my absence as well, so he could probably handle >>>this too. If you would like, I can probably spare a few minutes to >>>get this done today. However, the name "development crypt" sounds a >>>little, um, cryptic. Maybe we can find something to describe it a >>>little better? >>> >>>Garry W. Cook, CCNA >>>Network Infrastructure Manager >>>MACTEC, Inc. - http://www.mactec.com/ >>>303.273.5050 (Office) - 720.220.1862 (Mobile) >>> >>>Yves wrote: >>> >>> >>>>OK, good. >>>> >>>>Ben or Garry, could you make a page "development crypt" and >>>>put this and other good >>>>stuff that we sometimes write in mails and then forget ? >>>> >>>>------------------------------------------------------------ >>>><h1>How to use Nightly CVS Tarball</h1> >>>><pre> >>>>One possibility would be to use the "Nightly CVS Tarball" >>>>from sf. This >>>>can be found at >>>>http://cvs.sourceforge.net/cvstarballs/perfparse-cvsroot.tar.bz2 >>>>This tarball is a complete cvs repository including all historical >>>>versions. This should work: >>>>- untar it >>>>- use "cvs -d/path/where/untared/perfparse checkout >>>> perfparse-phpgui" to get the latest revisions. >>>></pre> >>>>------------------------------------------------------------ >>>>If you have time, I also suggest a wiki. One for the user FAQ >>>>And one where we could easily write such things instead of >>>>downloading the file, edit it, upload. >>>> >>>>Thanks Flo for the tip :) >>>> >>>>Yves > > > > > ------------------------------------------------------- > SF email is sponsored by - The IT Product Guide > Read honest & candid reviews on hundreds of IT Products from real users. > Discover which products truly live up to the hype. Start reading now. > http://productguide.itmanagersjournal.com/ > _______________________________________________ > Perfparse-devel-int mailing list > Per...@li... > https://lists.sourceforge.net/lists/listinfo/perfparse-devel-int > |
|
From: Cook, G. <GW...@ma...> - 2004-11-29 15:43:58
|
I've not used a WIKI/TIKI, but I hear that they are not too difficult to setup. Think it requires a backend DB, although not positive. Like you mentioned below, sf.net will give access to MySQL, but I believe we need to make a formal request for this. Garry W. Cook, CCNA Network Infrastructure Manager MACTEC, Inc. - http://www.mactec.com/ 303.273.5050 (Office) - 720.220.1862 (Mobile) Ben Clewett wrote: > Garry, >=20 > One thing that Yves talked about was a tiki system. Have you ever set > one up? This could be really useful for, as Yves suggested, an FAQ > and a private document of agreed bits. sf.net give access to the > database, but not looked any further at it... >=20 > Ben >=20 > Cook, Garry wrote: >=20 >> Sorry, I haven't been following this thread very closely, I'll have >> to go back and read it all the way through. However, to add a page, >> this is not too hard. I'm sure Flo can do it, as he has been working >> with PHP a bit, the pages are written in PHP. Ben has been >> maintaining them in my absence as well, so he could probably handle >> this too. If you would like, I can probably spare a few minutes to >> get this done today. However, the name "development crypt" sounds a >> little, um, cryptic. Maybe we can find something to describe it a >> little better?=20 >>=20 >> Garry W. Cook, CCNA >> Network Infrastructure Manager >> MACTEC, Inc. - http://www.mactec.com/ >> 303.273.5050 (Office) - 720.220.1862 (Mobile) >>=20 >> Yves wrote: >>=20 >>> OK, good. >>>=20 >>> Ben or Garry, could you make a page "development crypt" and >>> put this and other good >>> stuff that we sometimes write in mails and then forget ? >>>=20 >>> ------------------------------------------------------------ >>> <h1>How to use Nightly CVS Tarball</h1> >>> <pre> >>> One possibility would be to use the "Nightly CVS Tarball" >>> from sf. This >>> can be found at >>> http://cvs.sourceforge.net/cvstarballs/perfparse-cvsroot.tar.bz2 >>> This tarball is a complete cvs repository including all historical >>> versions. This should work:=20 >>> - untar it >>> - use "cvs -d/path/where/untared/perfparse checkout >>> perfparse-phpgui" to get the latest revisions. >>> </pre> >>> ------------------------------------------------------------ >>> If you have time, I also suggest a wiki. One for the user FAQ >>> And one where we could easily write such things instead of >>> downloading the file, edit it, upload. >>>=20 >>> Thanks Flo for the tip :) >>>=20 >>> Yves |
|
From: Yves <yme...@pe...> - 2004-11-29 14:44:01
|
> Yves,
> How much do I get if I told you I've found the solution?
31 more characters in the ChangeLog, and some public thanks in the announ=
ce of
pp-0.104.2 that may be released this week :)
> :)
:)
About the 31 characters, I suggest "Tim Wuyts <tim...@gm...>" :)
Do you agree ?
> OK, I admit, it seems odd, but I've encountered this next comment in a
> piece of Perl (yes, Perl) code:
> # Open the fifo read/write, not just read. This is necessary because
> # of the POSIX rules about using select() on named pipes when no writer=
s
> # are present. This is a very key step that is hard to find documentati=
on
> # about.
>
> (note the last sentence!)
I confirm...
My searchs showed me that there was an additionnal bug : before the 1st w=
rite to the
fifo, perfparse would have hanged on that open(fifo, O_RDONLY).
Using O_RDONLY |O_NONBLOCK fixes that bug, but select() still immediately=
returns.
Using O_RDWR works and O_NONBLOCK is not necessary for open() :)
> So I tried exactly that:
> change
> log_fd[i]->fd =3D open(filename+1, O_RDONLY);
> to
> log_fd[i]->fd =3D open(filename+1, O_RDWR);
>
> and that's it ! Works in 103.2 and 104.1
> (perfparsed 104.1 compiles, but doesn't write anyhting to the db...
> :(, but that's another problem I guess. More tomorrow... )
For the database, the problem is maybe dlopen() not working. A possible r=
eason is
libpp_storage_mysql not being in a PATH that ld.so knows. On GNU/Linux, e=
dit
/etc/ld.so.conf then run ldconfig. On other systems (including GNU/Linux =
if you don't
want to edit /etc/ld.so.conf) change your $LD_LIBRARY_PATH env var.
Could you tell me any error message that you have ? I'd like to find wher=
e I should stop
the execution of perfparse when there is a problem with the db.
> happy coding,
Thanks :)
And you, happy debugging :)
Yvse
> Tim
>
> On Fri, 26 Nov 2004 17:08:42 +0100 (CET), Yves <yme...@pe...>=
wrote:
>>
>> > Yves,
>> >
>> > Recompiling with the log_source_nonblock(i) line commented out does
>> > not change anything. The CPU still sky-rockets, and the 4 lines stil=
l
>> > show up (a gazillion times) in strace output.
>>
>> So I just suggest that you upgrade to 0.104.1 this week-end, when Ben =
releases it, and
>> I
>> will try to find out why there is always something to read on the pipe=
for select().
>>
>> The only thing I can suggest you now, if you really want a hacked solu=
tion, is to put
>> this in log_reader() :
>>
>> while(10 =3D=3D (r=3Dread(...))) {
>> ...
>> }
>> if((r>0) && (r<10)) {
>> ...
>> } else if(r =3D=3D 0) {
>> sleep(1);
>> }
>>
>> Add the 2 last lines to sleep the program when it reads nothing. The C=
PU should still
>> sky-rocket, but passengers will now have 1 second to get inside the ro=
cket :)
>>
>> I will try to find something better for 0.104.2 or 0.105 or course :)
>>
>> Yves
>> PS. uncomment back log_source_nonblock(i) because it may have side eff=
ects.
>>
>> --
>>
>>
>> - Homepage - http://ymettier.free.fr - http://www.logicacmg.com -
>> - GPG key - http://ymettier.free.fr/gpg.txt -
>> - Maitretarot - http://www.nongnu.org/maitretarot/ -
>> - Perfparse - http://perfparse.sf.net/ -
>>
>>
>
>
--=20
- Homepage - http://ymettier.free.fr - http://www.logicacmg.com -
- GPG key - http://ymettier.free.fr/gpg.txt -
- Maitretarot - http://www.nongnu.org/maitretarot/ -
- Perfparse - http://perfparse.sf.net/ -
|
|
From: Tim W. <tim...@gm...> - 2004-11-29 14:22:45
|
Yves,
How much do I get if I told you I've found the solution?
:)
OK, I admit, it seems odd, but I've encountered this next comment in a
piece of Perl (yes, Perl) code:
# Open the fifo read/write, not just read. This is necessary because
# of the POSIX rules about using select() on named pipes when no writers
# are present. This is a very key step that is hard to find documentation
# about.
(note the last sentence!)
So I tried exactly that:
change
log_fd[i]->fd = open(filename+1, O_RDONLY);
to
log_fd[i]->fd = open(filename+1, O_RDWR);
and that's it ! Works in 103.2 and 104.1
(perfparsed 104.1 compiles, but doesn't write anyhting to the db...
:(, but that's another problem I guess. More tomorrow... )
happy coding,
Tim
On Fri, 26 Nov 2004 17:08:42 +0100 (CET), Yves <yme...@pe...> wrote:
>
> > Yves,
> >
> > Recompiling with the log_source_nonblock(i) line commented out does
> > not change anything. The CPU still sky-rockets, and the 4 lines still
> > show up (a gazillion times) in strace output.
>
> So I just suggest that you upgrade to 0.104.1 this week-end, when Ben releases it, and I
> will try to find out why there is always something to read on the pipe for select().
>
> The only thing I can suggest you now, if you really want a hacked solution, is to put
> this in log_reader() :
>
> while(10 == (r=read(...))) {
> ...
> }
> if((r>0) && (r<10)) {
> ...
> } else if(r == 0) {
> sleep(1);
> }
>
> Add the 2 last lines to sleep the program when it reads nothing. The CPU should still
> sky-rocket, but passengers will now have 1 second to get inside the rocket :)
>
> I will try to find something better for 0.104.2 or 0.105 or course :)
>
> Yves
> PS. uncomment back log_source_nonblock(i) because it may have side effects.
>
> --
>
>
> - Homepage - http://ymettier.free.fr - http://www.logicacmg.com -
> - GPG key - http://ymettier.free.fr/gpg.txt -
> - Maitretarot - http://www.nongnu.org/maitretarot/ -
> - Perfparse - http://perfparse.sf.net/ -
>
>
|
|
From: Yves <yme...@pe...> - 2004-11-26 16:09:06
|
> Yves,
>
> Recompiling with the log_source_nonblock(i) line commented out does
> not change anything. The CPU still sky-rockets, and the 4 lines still
> show up (a gazillion times) in strace output.
So I just suggest that you upgrade to 0.104.1 this week-end, when Ben rel=
eases it, and I
will try to find out why there is always something to read on the pipe fo=
r select().
The only thing I can suggest you now, if you really want a hacked solutio=
n, is to put
this in log_reader() :
while(10 =3D=3D (r=3Dread(...))) {
...
}
if((r>0) && (r<10)) {
...
} else if(r =3D=3D 0) {
sleep(1);
}
Add the 2 last lines to sleep the program when it reads nothing. The CPU =
should still
sky-rocket, but passengers will now have 1 second to get inside the rocke=
t :)
I will try to find something better for 0.104.2 or 0.105 or course :)
Yves
PS. uncomment back log_source_nonblock(i) because it may have side effect=
s.
--=20
- Homepage - http://ymettier.free.fr - http://www.logicacmg.com -
- GPG key - http://ymettier.free.fr/gpg.txt -
- Maitretarot - http://www.nongnu.org/maitretarot/ -
- Perfparse - http://perfparse.sf.net/ -
|
|
From: Tim W. <tim...@gm...> - 2004-11-26 15:56:18
|
Yves,
Recompiling with the log_source_nonblock(i) line commented out does
not change anything. The CPU still sky-rockets, and the 4 lines still
show up (a gazillion times) in strace output.
:(
Tim
On Fri, 26 Nov 2004 15:56:42 +0100 (CET), Yves <yme...@pe...> wrote:
>
> > Yves,
> >
> > I've done an strace, results are attached. Changing &tm to NULL did
> > not change anything.
>
> Well, here is what your perfparsed does when doing nothing :
>
> select(6, [4 5], NULL, NULL, NULL) = 1 (in [4])
> rt_sigprocmask(SIG_BLOCK, [INT TERM], [], 8) = 0
> read(4, "", 10) = 0
> rt_sigprocmask(SIG_SETMASK, [], NULL, 8) = 0
> wait4(-1, NULL, WNOHANG, NULL) = -1 ECHILD (No child processes)
>
> In other words, there is always something to read on the 4th file descriptor (1st line),
> even if we see on the 3rd line that it reads nothing.
>
> There is an additionnal test that you could maybe do, still in log_reader.c. You can
> find in open_log_source a big test :
> if(!strcmp(filename,"-")) {
> ...
> } else if(filename[0] == '>') {
> ...
> } else if(filename[0] == '|') {
> ...
> log_source_nonblock(i);
> ...
> } else {
> ...
> }
>
> Could you comment that log_source_nonblock(i), and only that one ?
> I hope this is the fix. Otherwise, I will have to find why, when there is nothing to
> read on the pipe, why it considers there is something to read anyways.
>
> Note : per...@li... is the mailing-list for the core developers,
> including Ben. If you don't mind, answer with that list in CC :)
>
>
>
> Yves
>
> --
> - Homepage - http://ymettier.free.fr - http://www.logicacmg.com -
> - GPG key - http://ymettier.free.fr/gpg.txt -
> - Maitretarot - http://www.nongnu.org/maitretarot/ -
> - Perfparse - http://perfparse.sf.net/ -
>
>
|
|
From: Yves <yme...@pe...> - 2004-11-26 14:56:54
|
> Yves,
>
> I've done an strace, results are attached. Changing &tm to NULL did
> not change anything.
Well, here is what your perfparsed does when doing nothing :
select(6, [4 5], NULL, NULL, NULL) =3D 1 (in [4])
rt_sigprocmask(SIG_BLOCK, [INT TERM], [], 8) =3D 0
read(4, "", 10) =3D 0
rt_sigprocmask(SIG_SETMASK, [], NULL, 8) =3D 0
wait4(-1, NULL, WNOHANG, NULL) =3D -1 ECHILD (No child processes=
)
In other words, there is always something to read on the 4th file descrip=
tor (1st line),
even if we see on the 3rd line that it reads nothing.
There is an additionnal test that you could maybe do, still in log_reader=
.c. You can
find in open_log_source a big test :
if(!strcmp(filename,"-")) {
...
} else if(filename[0] =3D=3D '>') {
...
} else if(filename[0] =3D=3D '|') {
...
log_source_nonblock(i);
...
} else {
...
}
Could you comment that log_source_nonblock(i), and only that one ?
I hope this is the fix. Otherwise, I will have to find why, when there is=
nothing to
read on the pipe, why it considers there is something to read anyways.
Note : per...@li... is the mailing-list for the core =
developers,
including Ben. If you don't mind, answer with that list in CC :)
Yves
--=20
- Homepage - http://ymettier.free.fr - http://www.logicacmg.com -
- GPG key - http://ymettier.free.fr/gpg.txt -
- Maitretarot - http://www.nongnu.org/maitretarot/ -
- Perfparse - http://perfparse.sf.net/ -
|
|
From: Ben C. <BCl...@pe...> - 2004-11-23 08:39:58
|
Garry, One thing that Yves talked about was a tiki system. Have you ever set one up? This could be really useful for, as Yves suggested, an FAQ and a private document of agreed bits. sf.net give access to the database, but not looked any further at it... Ben Cook, Garry wrote: > Sorry, I haven't been following this thread very closely, I'll have to > go back and read it all the way through. However, to add a page, this is > not too hard. I'm sure Flo can do it, as he has been working with PHP a > bit, the pages are written in PHP. Ben has been maintaining them in my > absence as well, so he could probably handle this too. If you would > like, I can probably spare a few minutes to get this done today. > However, the name "development crypt" sounds a little, um, cryptic. > Maybe we can find something to describe it a little better? > > Garry W. Cook, CCNA > Network Infrastructure Manager > MACTEC, Inc. - http://www.mactec.com/ > 303.273.5050 (Office) - 720.220.1862 (Mobile) > > Yves wrote: > >>OK, good. >> >>Ben or Garry, could you make a page "development crypt" and >>put this and other good >>stuff that we sometimes write in mails and then forget ? >> >>------------------------------------------------------------ >><h1>How to use Nightly CVS Tarball</h1> >><pre> >>One possibility would be to use the "Nightly CVS Tarball" >>from sf. This >>can be found at >>http://cvs.sourceforge.net/cvstarballs/perfparse-cvsroot.tar.bz2 >>This tarball is a complete cvs repository including all historical >>versions. This should work: >>- untar it >>- use "cvs -d/path/where/untared/perfparse checkout perfparse-phpgui" >> to get the latest revisions. >></pre> >>------------------------------------------------------------ >>If you have time, I also suggest a wiki. One for the user FAQ >>And one where we could easily write such things instead of >>downloading the file, edit it, upload. >> >>Thanks Flo for the tip :) >> >>Yves > > > |
2 messages has been excluded from this view by a project administrator.