--> gemini has joined #net-snmp
--> df has joined #net-snmp
--> hampa has joined #net-snmp
[2002/10/08 03:05:42] #net-snmp <hampa> is it ok to increase the xxxEntry index?
[2002/10/08 06:33:21] #net-snmp <rstory> no, xxxEntry is *always* 1
--> wes has joined #net-snmp
--> nostaw has joined #net-snmp
[2002/10/08 06:46:16] #net-snmp <hampa> k.. thanks.. I will use 3 indexes instead
--> miike has joined #net-snmp
--> jbpn has joined #net-snmp
[2002/10/08 07:16:13] #net-snmp <jbpn> howdy
[2002/10/08 07:16:29] #net-snmp <rstory> how-do
[2002/10/08 07:29:49] #net-snmp <wes> ua: can you resend your perl patch using diff -u
--> MezoWork has joined #net-snmp
[2002/10/08 08:09:20] #net-snmp <uathome> wes: sure thing
--> TrogL has joined #net-snmp
[2002/10/08 08:52:48] #net-snmp <TrogL> either of you see the message from the guy in france? am I confused or is he trying to use net-snmp configuation and expect it to work with Sun's snmpd daemon?
[2002/10/08 08:53:32] #net-snmp <TrogL> I think what's happening is he's crashing Sun's daemon by sending it a request it doesnt' understand - why I sent to net-snmp in the first palce
[2002/10/08 08:53:43] #net-snmp <TrogL> s/sent/went/
[2002/10/08 08:54:14] #net-snmp <TrogL> s/either/any/ s/palce/place/ - grrrr
--> delanne_ has joined #net-snmp
[2002/10/08 09:14:15] #net-snmp <wes> I saw it, but didn'tunderstand it.
[2002/10/08 09:18:53] #net-snmp <TrogL> ok, i'll respond
[2002/10/08 09:26:53] #net-snmp <TrogL> argggg - a third person with the 100 disks problem
[2002/10/08 09:27:13] #net-snmp <TrogL> here's what I was hoping for as a solution,
[2002/10/08 09:27:27] #net-snmp <TrogL> find out somehow how many disks are on the system
[2002/10/08 09:27:34] #net-snmp <TrogL> create the arrays with that size
[2002/10/08 09:27:46] #net-snmp <TrogL> walk the tree, populating the arrays
[2002/10/08 09:28:06] #net-snmp <TrogL> output the results
[2002/10/08 09:28:26] #net-snmp <TrogL> my problem - the only way I've been able to find to get the disks ont he system is to walk the tree
[2002/10/08 09:28:30] #net-snmp <TrogL> chicken and egg
[2002/10/08 09:28:58] #net-snmp <TrogL> given there's now 3 of us having trouble, should I temporarily assign 0-9 for RAM and swap, 10-210 for disks?
[2002/10/08 09:29:12] #net-snmp <TrogL> I was hoping Sun would get back to me on kstat stuff
[2002/10/08 09:30:57] #net-snmp <rstory> why don't you try extracting the code that snmp uses to find disks into a utility?
[2002/10/08 09:31:44] #net-snmp <rstory> even better, just make it dynamic..
[2002/10/08 09:31:53] #net-snmp <rstory> count disks at startup and allocate array
[2002/10/08 09:33:54] #net-snmp <jbpn> or just keep reallocing the array as you walk
[2002/10/08 09:34:14] #net-snmp <jbpn> (but add, like, 10 each time)
[2002/10/08 09:35:10] #net-snmp <TrogL> aieeee....C's like my fifth language, I'm very weak on C memory management
[2002/10/08 09:35:17] #net-snmp <TrogL> I don't know how to code reallocating an array
[2002/10/08 09:35:23] #net-snmp <jbpn> man realloc
[2002/10/08 09:35:27] #net-snmp <TrogL> I did have the idea of counting them, then allocating, but I was worried about performance
[2002/10/08 09:35:47] #net-snmp <jbpn> that's not good enough anyway, because the number can change at run time
[2002/10/08 09:36:48] #net-snmp <rstory> make it a conf file token diskTableMaxSize
[2002/10/08 09:37:25] #net-snmp <TrogL> I'd prefer not to hard code on configure
[2002/10/08 09:37:49] #net-snmp <rstory> conf file = snmpd.conf = runtime
[2002/10/08 09:38:09] #net-snmp <jbpn> No no no
[2002/10/08 09:38:12] #net-snmp <TrogL> my problem is with a CD jukebox - they keep adding cd's
[2002/10/08 09:38:20] #net-snmp <TrogL> ACTION looks confused
[2002/10/08 09:38:20] #net-snmp <jbpn> The number of disks can change while snmpd is running.
[2002/10/08 09:39:05] #net-snmp <jbpn> It needs to be able to cope with this without being restarted.
[2002/10/08 09:39:11] #net-snmp <rstory> so pick a nice high max value...
[2002/10/08 09:39:20] #net-snmp <jbpn> Which doesn't leave many options...
[2002/10/08 09:39:22] #net-snmp <rstory> HUP should re-read configs..
[2002/10/08 09:39:51] #net-snmp <TrogL> I'd prefer not go to the high max route - memory bloat
[2002/10/08 09:40:05] #net-snmp <jbpn> Yes but how does putting a CD in a drive trigger snmpd getting HUPped?
[2002/10/08 09:40:56] #net-snmp <rstory> if your changer holds 50 cds, set max to 55.. if you buy a new changer, then you have to HUP manually
[2002/10/08 09:41:17] #net-snmp <rstory> I agree dynamic allocation is the *right* way
[2002/10/08 09:41:20] #net-snmp <TrogL> don't forget the guys with SANs - they reconfigure on the fly
[2002/10/08 09:41:50] #net-snmp <rstory> but if TrogL isn't comfortable with his C skills, a way to specifiy max at runtime is the next best thing
[2002/10/08 09:41:52] #net-snmp <jbpn> Even quite normal systems using NFS change number of filesystems dynamically
[2002/10/08 09:42:15] #net-snmp <TrogL> tahts' a different OID
[2002/10/08 09:42:29] #net-snmp <jbpn> Ooops. Beg pardon.
[2002/10/08 09:42:31] #net-snmp <TrogL> we're only reporting hard mounts as far as I know
[2002/10/08 09:43:33] #net-snmp <TrogL> I hate to bring this up, what about a snmpd command-line option for people with large #'s of disks?
[2002/10/08 09:43:53] #net-snmp <jbpn> That's almost what Robert is suggesting.
[2002/10/08 09:44:05] #net-snmp <jbpn> Except it's a token in snmpd.conf rather than a command line option.
[2002/10/08 09:44:16] #net-snmp <jbpn> (the namespace for the latter is already rather crowded)
[2002/10/08 09:44:29] #net-snmp <jbpn> Although, actually the two are sort of equivalent anyway.
[2002/10/08 09:44:57] #net-snmp <jbpn> Since --longOption is the same as putting longOption in snmpd.conf I believe.
[2002/10/08 09:45:29] #net-snmp <TrogL> aren't there two snmpd.conf's - one you fiddle with and one dynamic?
[2002/10/08 09:45:45] #net-snmp <jbpn> Yes (I meant the one you fiddle with)
[2002/10/08 09:46:13] #net-snmp <TrogL> I could go for that -for a variety of reasons;
[2002/10/08 09:46:25] #net-snmp <jbpn> The realloc stuff isn't much harder...
[2002/10/08 09:46:25] #net-snmp <TrogL> it's probably within my skill set
[2002/10/08 09:46:32] #net-snmp <TrogL> I don't have to run the tree twice
[2002/10/08 09:46:32] #net-snmp <jbpn> honest!
[2002/10/08 09:47:19] #net-snmp <TrogL> ACTION mans realloc
[2002/10/08 10:02:44] #net-snmp <TrogL> ok this doesn't look too bad - my boss is encouraging me to use calloc over malloc when originally defining the arrays
[2002/10/08 10:02:46] #net-snmp <TrogL> is this a good thing?
[2002/10/08 10:02:54] #net-snmp <jbpn> Usually, yes.
[2002/10/08 10:08:02] #net-snmp <TrogL> ACTION cvs's
[2002/10/08 10:08:15] #net-snmp <jbpn> bye for now
[2002/10/08 10:33:51] #net-snmp <TrogL> sun's finally gonna send me stuff on kstat
--> delanne__ has joined #net-snmp
[2002/10/08 10:42:12] #net-snmp <uathome> morning all
[2002/10/08 10:45:14] #net-snmp <TrogL> mrng
--> benr has joined #net-snmp
--> delanne_ has joined #net-snmp
--> niaht_ has joined #net-snmp
--> miike has joined #net-snmp
--> benr has joined #net-snmp
[2002/10/08 15:44:06] #net-snmp <uathome> I tried those tests on my home linux box today. The 100+ SNMP tests have similar failures. OID doesn't fail though.
[2002/10/08 15:44:06] #net-snmp <uathome> Seems as if the Sparc isn't compiling sysORTable, for whatever reason, while the Linux box is.
--> AmateurSurgeon has joined #net-snmp
[2002/10/08 15:45:48] #net-snmp <AmateurSurgeon> Hi, i've got a quick question about the snmp-lib - is anyone around ? :)
[2002/10/08 15:46:03] #net-snmp <uathome> I'm here, shoot.
[2002/10/08 15:47:15] #net-snmp <AmateurSurgeon> I'm planning to do a final year uni project which involes collecting snmp data and logging it to a database, then doing cool things with the data. Obviously, i'm going to have to do a lot of the grunt work myself to get marks, but am i on the right track by wanting to use the snmp libs as a starting point ?
[2002/10/08 15:48:26] #net-snmp <uathome> You mean, using the net-snmp libraries, right?
[2002/10/08 15:48:28] #net-snmp <AmateurSurgeon> i need to find that threshold between not making it too easy, and not having to do so much work that i find myself 10% through with a week to go :)
[2002/10/08 15:48:29] #net-snmp <AmateurSurgeon> yep
[2002/10/08 15:48:35] #net-snmp <uathome> Sure thing, that's what they're for.
[2002/10/08 15:48:48] #net-snmp <AmateurSurgeon> they seem like a good basis without doing too much of the work for me
[2002/10/08 15:48:52] #net-snmp <AmateurSurgeon> cool, thanks !
[2002/10/08 15:49:29] #net-snmp <AmateurSurgeon> ttfn
[2002/10/08 16:03:45] #net-snmp <wes> uathome: try running "snmptranslate -IR sysORTable" on your solaris box.
[2002/10/08 16:07:58] #net-snmp <uathome> Did not find 'zeroDotZero' in module SNMPv2-SMI (/usr/local/share/snmp/mibs/DISMAN-SCHEDULE-MIB.txt)
[2002/10/08 16:07:58] #net-snmp <uathome> Unknown object identifier: sysORTable
[2002/10/08 16:08:49] #net-snmp <uathome> bbiab, need to buy some thermal CPU grease before the store closes.
[2002/10/08 16:11:24] #net-snmp <wes> odd.
[2002/10/08 16:11:39] #net-snmp <wes> Um, do you have MIBS set to "all" or something?
--> rstory has joined #net-snmp
--> gemini has joined #net-snmp
[2002/10/08 21:18:08] #net-snmp <gemini> hello guys
[2002/10/08 21:18:56] #net-snmp <gemini> i am confused now what to put in snmp.conf and snmpd.conf
[2002/10/08 21:19:16] #net-snmp <gemini> how can i distinguish?
[2002/10/08 21:19:36] #net-snmp <gemini> HELP!
[2002/10/08 21:19:57] #net-snmp <rstory> snmpd -H
[2002/10/08 21:20:42] #net-snmp <rstory> snmpd.conf is for the agent... snmp.conf is for the apps... (more or less)
[2002/10/08 21:22:02] #net-snmp <gemini> those security level stuff goes into snmp.conf?
[2002/10/08 21:22:28] #net-snmp <rstory> default security levels? yes
[2002/10/08 21:22:43] #net-snmp <gemini> ok....
--> delanne__ has joined #net-snmp
--> jbpn has joined #net-snmp
[2002/10/08 23:55:08] #net-snmp <gemini> ok what does this mean... someone plz help
[2002/10/08 23:55:26] #net-snmp <gemini> i put a community name "test" in my snmp.conf file
[2002/10/08 23:55:29] #net-snmp <gemini> and tried
[2002/10/08 23:55:42] #net-snmp <gemini> snmpget -c test localhost system.sysUpTime.0
[2002/10/08 23:56:12] #net-snmp <gemini> and it said
[2002/10/08 23:56:13] #net-snmp <gemini> Error in packet
[2002/10/08 23:56:13] #net-snmp <gemini> Reason: (noSuchName) There is no such variable name in this MIB.
[2002/10/08 23:56:13] #net-snmp <gemini> Failed object: SNMPv2-MIB::sysUpTime.0
[2002/10/08 23:56:46] #net-snmp <gemini> but when i tried "snmpget -c public localhost system.sysUpTime.0"
[2002/10/08 23:56:54] #net-snmp <gemini> it gave me a correct response
[2002/10/08 23:57:01] #net-snmp <gemini> what is wrong??
[2002/10/08 23:57:39] #net-snmp <gemini> anyone around ????

Powered by: