--> linux119 has joined #net-snmp
[2002/10/16 01:21:34] #net-snmp <linux119> hi
--> gemini has joined #net-snmp
[2002/10/16 02:11:01] #net-snmp <gemini> hello
[2002/10/16 02:11:17] #net-snmp <gemini> why can't my compiler find the net-snmp library??
[2002/10/16 02:11:31] #net-snmp <gemini> make snmpdemoapp
[2002/10/16 02:11:46] #net-snmp <gemini> cc -o snmpdemoapp snmpdemoapp.o `net-snmp-config --libs`
[2002/10/16 02:11:46] #net-snmp <gemini> ld: fatal: library -lnetsnmp: not found
[2002/10/16 02:11:46] #net-snmp <gemini> ld: fatal: File processing errors. No output written to snmpdemoapp
[2002/10/16 02:11:46] #net-snmp <gemini> make: *** [snmpdemoapp] Error 1
[2002/10/16 02:11:48] #net-snmp <gemini> any help??
--> guest has joined #net-snmp
[2002/10/16 02:22:42] #net-snmp <guest> hi,everyone.
[2002/10/16 02:23:12] #net-snmp <guest> who used the parameter about snmptrapd -l?
--> ajcl has joined #net-snmp
[2002/10/16 03:11:31] #net-snmp <ajcl> hello where can i find documentation for programming snmp ?
[2002/10/16 04:05:02] #net-snmp <hampa> www.net-snmp.org
[2002/10/16 05:01:01] #net-snmp <hampa> ACTION is away: I'm busy
[2002/10/16 05:50:33] #net-snmp <hampa> ACTION is back (gone 00:49:31)
[2002/10/16 07:11:29] #net-snmp <hampa> what is the purpose of the last .0 in a scalar?
[2002/10/16 07:11:49] #net-snmp <hampa> could it be used as an index or should it always be .0 ?
[2002/10/16 07:36:13] #net-snmp <rstory> hampa: a scalar *always* ends with .0
[2002/10/16 07:36:46] #net-snmp <rstory> I'm not sure what you mean by using it as an index
[2002/10/16 07:36:48] #net-snmp <TrogL> try '[snmp] mibs all'?
[2002/10/16 07:37:53] #net-snmp <hampa> myval.5 instead of myval.0
[2002/10/16 07:38:11] #net-snmp <TrogL> in /usr/local/share/snmp/snmpd.conf ?
[2002/10/16 07:38:46] #net-snmp <rstory> gemini: did you run 'make install' after you built the source?
[2002/10/16 07:39:20] #net-snmp <rstory> hampa: nope, can't do that. if you need more than one value for myval, you need a table.
[2002/10/16 07:40:06] #net-snmp <rstory> TrogL: yes, that should work
[2002/10/16 07:40:10] #net-snmp <hampa> k.. just wondering
[2002/10/16 07:40:32] #net-snmp <hampa> another question.. should tables indexes begin on 0 or 1 ?
[2002/10/16 07:43:14] #net-snmp <TrogL> what is the sysservices object and why is it set to 79?
[2002/10/16 07:47:50] #net-snmp <rstory> hampa: i definitely recommend starting at 1, but I'm not sure if it is actually required
[2002/10/16 07:49:08] #net-snmp <rstory> TrogL: see the mib definition... or run snmpconf -g basic_setup, which will ask you questions to set a different value
[2002/10/16 07:52:10] #net-snmp <hampa> I have implemented a table handler in c.. Is it possible to make it so that it starts asking for index 1 when I am doing a walk?
[2002/10/16 08:05:36] #net-snmp <rstory> not on the agent side.. the agent passes on whatever snmpwalk sends... so you can test with a specific snmpwalk command line, but you can't ever depend on the index being a specific value on the agent
--> miike has joined #net-snmp
[2002/10/16 08:13:52] #net-snmp <hampa> k
--> li_nu_xx has joined #net-snmp
--> PerlStalker has joined #net-snmp
--> ua has joined #net-snmp
[2002/10/16 09:44:05] #net-snmp <PerlStalker> Can someone, please, point me to some docs that explain how to set which syslog faclity to use?
[2002/10/16 10:21:46] #net-snmp <TrogL> a users-digest article suggests that there is no guarantee taht indexes will remain the same from one moment to the next
[2002/10/16 10:21:58] #net-snmp <TrogL> they're talking specifically about netowrk appearing and disappearing
[2002/10/16 10:22:35] #net-snmp <TrogL> could I use the same concept with my RAM/swap/filesystem woes? it's pretty much what I've been thinking all alnog, but I dont' want to disappoint
[2002/10/16 10:58:28] #net-snmp <PerlStalker> ACTION is away: Lunch
[2002/10/16 11:04:45] #net-snmp <TrogL> is theer a way to completely remove net-snmp from my system and start over? - how can it NOT handle SNMPv1 correctly?
[2002/10/16 11:05:19] #net-snmp <rstory> depends on the mib... for example, the host mib states that hrDeviceIndex must be constant during the lifetime of the agent
[2002/10/16 11:05:46] #net-snmp <rstory> no, we don't have a make uninstall... but a fresh make install should overwrite (assuming you didn't configure different prefixes)
[2002/10/16 11:09:07] #net-snmp <TrogL> damn, hrDeviceIndex was the one that's causing me grief .. and you're right it goes all the way back to RFC1415
[2002/10/16 11:09:58] #net-snmp <TrogL> anyway, I'm baffled. I added agentx to my configure line, v1 still fails as do all the agent tests - on two different boxes
[2002/10/16 11:10:14] #net-snmp <TrogL> I can't proceed with my ORACLE or hosts work until this is fixed
[2002/10/16 11:10:44] #net-snmp <TrogL> do the test record anywhere WHY they failed?
[2002/10/16 11:10:59] #net-snmp <TrogL> I read through the test scripts and they do multiple things
[2002/10/16 11:13:58] #net-snmp <rstory> there should be some directories in /tmp
[2002/10/16 11:15:58] #net-snmp <TrogL> foundem
[2002/10/16 11:18:30] #net-snmp <TrogL> hmmm... notavalidcommunity
--> TrogL has joined #net-snmp
--> rstory has joined #net-snmp
[2002/10/16 11:22:33] #net-snmp <TrogL> why is the tester using port 8765?
[2002/10/16 11:23:05] #net-snmp <rstory> so it doesn't interfere with any agent running on 161...
[2002/10/16 11:23:18] #net-snmp <rstory> it runs with a specific config on a specific port...
[2002/10/16 11:23:23] #net-snmp <TrogL> ah, so my shutting the agent down during testing is unnecessary?
[2002/10/16 11:23:27] #net-snmp <rstory> a nice controlled envrionment
[2002/10/16 11:23:31] #net-snmp <rstory> yes..
[2002/10/16 11:24:39] #net-snmp <TrogL> my test is asking to get DISMAN-EVENT-MIB::sysUpTimeInstance and keeps getting told "notavalidcommunity"
[2002/10/16 11:24:41] #net-snmp <TrogL> WTF?
[2002/10/16 11:27:50] #net-snmp <rstory> what do you mean 'your test'?? a different test other then make test?
[2002/10/16 11:28:10] #net-snmp <TrogL> sorry, "the test"
[2002/10/16 11:42:28] #net-snmp <TrogL> it just does not seem to be handling DISMAN-EVENT-MIB::sysUpTimeInstance well
[2002/10/16 11:45:15] #net-snmp <slam> do i have to specify an engineID to send snmptrap even when the security level is set to noAuthNopriv?
[2002/10/16 11:55:18] #net-snmp <PerlStalker> ACTION is back (gone 00:56:49)
--> root has joined #net-snmp
[2002/10/16 12:33:11] #net-snmp <root> Hello all
[2002/10/16 12:33:31] #net-snmp <root> Hum...
[2002/10/16 12:33:34] #net-snmp <root> Wrong user...
--> beta2K has joined #net-snmp
[2002/10/16 12:35:29] #net-snmp <beta2K> That's better
[2002/10/16 12:35:43] #net-snmp <beta2K> Could someone explain something to me?
[2002/10/16 12:35:52] #net-snmp <beta2K> I'm a little convused about MIBs
[2002/10/16 12:41:04] #net-snmp <TrogL> ACTION will try
[2002/10/16 12:41:13] #net-snmp <beta2K> Cool :)
[2002/10/16 12:41:38] #net-snmp <PerlStalker> Can someone, please, point me to some docs that explain how to set which syslog faclity to use for snmpd?
[2002/10/16 12:42:15] #net-snmp <beta2K> Here's my problem, how come I look through the MIB files, I find a MIB I want to use, I request it from snmpd and it tells me it's not vaild.
[2002/10/16 12:45:33] #net-snmp <beta2K> Eg, I'm trying to graph disk usage, I tried snmpwalk -v 1 -c PUBLIC localhost hrStorageSize.1, it tells me the MIB is invalid
[2002/10/16 12:48:32] #net-snmp <rstory> the agent must support the mib... the presence of a mib does not imply support for the mib...
[2002/10/16 12:48:56] #net-snmp <beta2K> Ok, so how would I go about getting snmpd to support the mib?
[2002/10/16 12:48:59] #net-snmp <rstory> if the agent is the net-snmp agent, it needs to be configured with "--with-mib-modules=host" to add host support
[2002/10/16 12:49:11] #net-snmp <beta2K> Ok :)
[2002/10/16 12:49:21] #net-snmp <rstory> even them support varies by platform...
[2002/10/16 12:49:26] #net-snmp <beta2K> Is there a list of the modules for net-snmp?
[2002/10/16 12:49:42] #net-snmp <rstory> ./configure --help lists the optional modules
[2002/10/16 12:49:52] #net-snmp <beta2K> Thanks
[2002/10/16 12:49:57] #net-snmp <beta2K> *PooF*
[2002/10/16 12:53:50] #net-snmp <TrogL> rstory any ideas whatsoever on diagnosing why I sudenly lost support for v1? or should I just pull everything and start over? it's happening on two boxes
[2002/10/16 13:06:46] #net-snmp <PerlStalker> How can I get snmpd to rotate the log file or start a new one if I rotate it?
[2002/10/16 13:13:01] #net-snmp <TrogL> it starts a new one by default
[2002/10/16 13:13:06] #net-snmp <TrogL> -A appends instead
[2002/10/16 13:13:55] #net-snmp <TrogL> if you want it to rotate (eg. like /var/adm/messages?) you'll need to write something
[2002/10/16 13:15:55] #net-snmp <rstory> TrogL: run the agent with more debug flags.. are you specifying -v1 on the cmd line?
[2002/10/16 13:16:56] #net-snmp <TrogL> i'm allowing default of v3 in ./configure - i'm doing a make test and it's failing
[2002/10/16 13:17:40] #net-snmp <TrogL> the only thing that's changed on the system is I was trying to use it with ORACLE, and I hadn't updated my CVS in a week
[2002/10/16 13:18:02] #net-snmp <rstory> hmm... it's failing for me too..
[2002/10/16 13:18:08] #net-snmp <rstory> I'll take a look at it..
[2002/10/16 13:18:16] #net-snmp <TrogL> good, now I don't feel so bad
[2002/10/16 13:18:36] #net-snmp <TrogL> I have a suspicion...
[2002/10/16 13:18:56] #net-snmp <TrogL> DISMAN-EVENT-MIB::sysUpTimeInstance
[2002/10/16 13:18:59] #net-snmp <TrogL> is needed by several tests
[2002/10/16 13:19:03] #net-snmp <TrogL> if that is failing...
[2002/10/16 13:19:18] #net-snmp <PerlStalker> TrogL: rotation isn't a problem but snmpd doesn't create a new file once the old one is moved.
[2002/10/16 13:19:51] #net-snmp <TrogL> if the file is moved while snmpd is running, the OS will keep the inode
[2002/10/16 13:20:09] #net-snmp <TrogL> you have to stop snmpd, move the file, restart
[2002/10/16 13:20:35] #net-snmp <TrogL> you could use the -l option and create a new file with its own timestamp and save yourself the bother
[2002/10/16 13:20:37] #net-snmp <TrogL> of rotating
[2002/10/16 13:21:30] #net-snmp <TrogL> granted, you would need to root out the oldies, or maybe your script could rotate fielnames
[2002/10/16 13:21:32] #net-snmp <TrogL> then use -l to create them
[2002/10/16 13:22:17] #net-snmp <PerlStalker> TrogL: I could but the problem is that I don't plan to restart these boxes too often and I don't want to have to restart snmpd every night or so just to move the log file.
[2002/10/16 13:23:11] #net-snmp <PerlStalker> I guess I was hoping I could send it a SIGHUP or SIGUSR1 to get it to reopen the log file.
[2002/10/16 13:23:21] #net-snmp <rstory> use -s option to log to syslog
[2002/10/16 13:23:42] #net-snmp <PerlStalker> rstory: How do you tell which facility to use?
[2002/10/16 13:23:59] #net-snmp <PerlStalker> or does it use LOCAL0?
[2002/10/16 13:26:12] #net-snmp <rstory> looks like it uses LOG_DAEMON
[2002/10/16 13:26:32] #net-snmp <PerlStalker> Ok
[2002/10/16 13:26:55] #net-snmp <rstory> wouldn't be to hard to patch it to use another though... snmptrapd already has code for changing facility at runtime
[2002/10/16 13:27:28] #net-snmp <PerlStalker> rstory: That might be more trouble than it's worth (for me).
[2002/10/16 13:28:58] #net-snmp <PerlStalker> Yup. LOG_DAEMON. Thanks.
[2002/10/16 13:41:00] #net-snmp <rstory> TrogL: aha... v1 works, but the agent is ignoring SIGTERM / Ctrl-C !?!?
[2002/10/16 13:41:47] #net-snmp <TrogL> it's SUPPOED to ignore ctrl-c!
[2002/10/16 13:42:17] #net-snmp <TrogL> othrwise the next ctrl-c you type in the session crashed the daemon
[2002/10/16 13:42:18] #net-snmp <rstory> hmm... I"m not sure I agree..
[2002/10/16 13:42:31] #net-snmp <TrogL> wes and I went around this about a month ago
[2002/10/16 13:42:54] #net-snmp <TrogL> send it an INT
[2002/10/16 13:44:08] #net-snmp <TrogL> whatever broke, broke sometime in the last week
[2002/10/16 13:52:12] #net-snmp <rstory> got it... fix going into CVS in a minute or so..
--> benr has joined #net-snmp
[2002/10/16 13:52:31] #net-snmp <TrogL> and it was....
[2002/10/16 13:52:45] #net-snmp <TrogL> is the mib.c fix there?
[2002/10/16 13:55:38] #net-snmp <TrogL> lemme know when I can CVS
[2002/10/16 14:06:38] #net-snmp <rstory> hmm.. fixed the original problem, but there appears to be another...
[2002/10/16 14:07:16] #net-snmp <TrogL> need help?
[2002/10/16 14:11:51] #net-snmp <rstory> nope, I think I've got it under control...
[2002/10/16 14:14:37] #net-snmp <TrogL> ACTION is glad to be of help....or such a pain in the ass digging up "features"
[2002/10/16 14:15:47] #net-snmp <rstory> ok, checking in fix now...
[2002/10/16 14:16:50] #net-snmp <rstory> a recent change tried to reselect after a signal.. however, this didn't allow the agent to see the flags indicating the need to act on the signal..
[2002/10/16 14:17:14] #net-snmp <rstory> thus it never exited on SIGTERM, causing the test to fail...
[2002/10/16 14:18:37] #net-snmp <TrogL> ah!
[2002/10/16 14:18:54] #net-snmp <TrogL> solaris only?
[2002/10/16 14:19:07] #net-snmp <rstory> geez, sf cvs is slow today!
[2002/10/16 14:19:12] #net-snmp <rstory> no, all platforms
[2002/10/16 14:19:32] #net-snmp <TrogL> ACTION looks pleased with himself
[2002/10/16 14:19:55] #net-snmp <rstory> ok, checkin finished... update away
[2002/10/16 14:20:54] #net-snmp <TrogL> will removing my temporary mib.c file cause it to update correctly, or will that throw things out of whack?
[2002/10/16 14:21:27] #net-snmp <TrogL> last time I left a "fix" there, it got pissy
[2002/10/16 14:21:28] #net-snmp <rstory> yes, if you don't want to keep your changes..
[2002/10/16 14:21:44] #net-snmp <rstory> should work both ways
[2002/10/16 14:21:56] #net-snmp <rstory> unless your fix conflicts with something in the newer version
[2002/10/16 14:22:31] #net-snmp <TrogL> it's "your" fix! - from yesterday
[2002/10/16 14:25:03] #net-snmp <TrogL> I got new: snmp_agent.c, snmpd.c, snmptrapd.c, snmptrapd_log.c+h and mib.c
[2002/10/16 14:26:38] #net-snmp <rstory> looks about right
[2002/10/16 14:32:45] #net-snmp <TrogL> does this look right...?
[2002/10/16 14:33:32] #net-snmp <TrogL> ./configure --with-mib-modules="host disman/event-mib ucd-snmp/diskio smux agent" --with-cc=gcc --enable-shared --enable-embedded-perl
[2002/10/16 14:34:20] #net-snmp <rstory> did you mean for agent to be agentx?
[2002/10/16 14:34:48] #net-snmp <TrogL> probably
[2002/10/16 14:34:51] #net-snmp <TrogL> yep
[2002/10/16 14:35:41] #net-snmp <TrogL> ORACLE needs to access it as an agent
[2002/10/16 14:42:28] #net-snmp <TrogL> ACTION makes on 3 platforms
[2002/10/16 14:56:46] #net-snmp <TrogL> good test on 2_6 except 15 failed - v3 user creation and usage using snmpusm
[2002/10/16 14:57:58] #net-snmp <TrogL> "encryption support not enabled"
[2002/10/16 14:58:04] #net-snmp <TrogL> "snmpget: USM encryption error"
[2002/10/16 14:58:26] #net-snmp <TrogL> this doesn't entirely surprise me as I'm unsure exactly what libraries are installed on this particular box
[2002/10/16 14:58:36] #net-snmp <rstory> yep... probably missing openssl
[2002/10/16 14:58:59] #net-snmp <TrogL> however, I did do a make distclean then ./configure so it should have figured it out on its own
[2002/10/16 14:59:10] #net-snmp <TrogL> we'll see what happens on the other two
[2002/10/16 14:59:34] #net-snmp <TrogL> one due up shortly, the other will be sometime tomorrow (Sparc 5)
[2002/10/16 15:08:00] #net-snmp <rstory> configure doesn't tell the test system it's missing... a pet peeve of mine...
[2002/10/16 15:08:15] #net-snmp <TrogL> was just about to bitch and complain about that...just discovered same thing
[2002/10/16 15:10:03] #net-snmp <TrogL> isn't it kind of poitnless to default to v3 if you don't have opensll?
[2002/10/16 15:10:41] #net-snmp <rstory> no, as we have internal AUTH methods...
[2002/10/16 15:10:53] #net-snmp <TrogL> k
[2002/10/16 15:27:43] #net-snmp <TrogL> would it ever occur to have two copies of snmpd running at same time (ie. in a meaningful manner)?
[2002/10/16 15:28:34] #net-snmp <rstory> i'm not sure what you mean by meaningful manner, but yes, it could happen
[2002/10/16 15:29:15] #net-snmp <TrogL> is it a good thing? should we be having a test to see if it's already running? I just got caught
[2002/10/16 15:32:32] #net-snmp <rstory> caught how?
[2002/10/16 15:32:43] #net-snmp <rstory> there are probably some gotchas that we don't handle well..
[2002/10/16 15:32:44] #net-snmp <TrogL> had two copies running
[2002/10/16 15:32:52] #net-snmp <TrogL> the older version was answering instead of the new one
[2002/10/16 15:33:01] #net-snmp <rstory> but only on different ports...
[2002/10/16 15:33:06] #net-snmp <TrogL> only way I caught it was the string that says its status
[2002/10/16 15:33:27] #net-snmp <TrogL> didn't say "smux" so I knew it was the wrong one
[2002/10/16 15:33:55] #net-snmp <rstory> an attempt to run the agent without a port specified twice should fail on the second...
[2002/10/16 15:34:39] #net-snmp <TrogL> nope - two running
[2002/10/16 15:35:40] #net-snmp <rstory> not possible... unix won't let you open the same port twice...
[2002/10/16 15:36:12] #net-snmp <TrogL> ps -ef showed me two running
[2002/10/16 15:36:26] #net-snmp <TrogL> and both kicked out traps
[2002/10/16 15:37:13] #net-snmp <rstory> one of them had to be on a different port (maybe a holdover from a make test run)
[2002/10/16 15:37:29] #net-snmp <TrogL> lemme try again
[2002/10/16 15:38:22] #net-snmp <TrogL> ps -ef reports no snmpd
[2002/10/16 15:38:49] #net-snmp <TrogL> when I do /usr/local/sbin/snmpd
[2002/10/16 15:38:57] #net-snmp <TrogL> it starts and kicks out traps
[2002/10/16 15:39:32] #net-snmp <TrogL> it's PID 5857
[2002/10/16 15:40:01] #net-snmp <TrogL> snmpwalk returned stuff
[2002/10/16 15:40:22] #net-snmp <TrogL> another /usr/local/sbin/snmpd
[2002/10/16 15:41:03] #net-snmp <TrogL> more traps
[2002/10/16 15:41:23] #net-snmp <TrogL> now have two PID's 5837 and 5913
[2002/10/16 15:41:41] #net-snmp <TrogL> snmpwalk returns stuff
[2002/10/16 15:42:01] #net-snmp <TrogL> my suspicion, based on earlier evidence, is that the first copy retains the port
[2002/10/16 15:42:09] #net-snmp <TrogL> because I was getting older information
[2002/10/16 15:42:13] #net-snmp <rstory> well color me stupid... you're right..
[2002/10/16 15:42:28] #net-snmp <rstory> but on my box, the most recent gets the port
[2002/10/16 15:43:11] #net-snmp <TrogL> or worse, they're both answering at random
[2002/10/16 15:43:26] #net-snmp <TrogL> which as you said, should be imposible
[2002/10/16 15:44:30] #net-snmp <TrogL> wonder if the higher PID gets the port?
[2002/10/16 15:44:37] #net-snmp <TrogL> or lower
[2002/10/16 15:45:31] #net-snmp <rstory> it is definitely the most recent on my box.. and not alternating or anything else...
[2002/10/16 15:45:48] #net-snmp <rstory> this turns my understanding of sockets upside down!
[2002/10/16 15:46:03] #net-snmp <TrogL> you can read multiple tiems fromt he ame socket
[2002/10/16 15:46:08] #net-snmp <TrogL> I've got another app runs on taht basis
[2002/10/16 15:46:49] #net-snmp <TrogL> but as far as I can remember, you can't open a socket twice to write
[2002/10/16 15:48:21] #net-snmp <rstory> hmm... seems to have been introduced in 5.0.3... 5.0.1 won't run with an existing agent...
[2002/10/16 15:49:11] #net-snmp <TrogL> ACTION screams with excitement
[2002/10/16 15:49:27] #net-snmp <TrogL> I GOT ORACLE STUFF!!
[2002/10/16 15:52:06] #net-snmp <TrogL> i can see the databases, the instances, something's down...
[2002/10/16 15:53:23] #net-snmp <TrogL> agentx and/or smux seems to have made the difference
[2002/10/16 15:53:44] #net-snmp <TrogL> that needs to go into a doc
[2002/10/16 15:53:50] #net-snmp <TrogL> do we need a README.ORACLE?
[2002/10/16 15:57:00] #net-snmp <TrogL> the document
[2002/10/16 15:57:01] #net-snmp <TrogL> http://sourceforge.net/docman/display_doc.php?docid=5201&group_id=12694
[2002/10/16 15:57:03] #net-snmp <TrogL> is out of date
[2002/10/16 15:57:31] #net-snmp <TrogL> written for version 4 and there's changed for ORACLE 9i
[2002/10/16 16:01:16] #net-snmp <TrogL> ORACLE's SNMP handled differently
[2002/10/16 16:08:19] #net-snmp <TrogL> ACTION thinks he found an ORACLE bug
[2002/10/16 16:09:14] #net-snmp <TrogL> hmmm - here's my problem
[2002/10/16 16:09:22] #net-snmp <TrogL> it runs fine if I have everything in "public"
[2002/10/16 16:09:48] #net-snmp <TrogL> if I switch everything to "mydomain" it kicks out a bazillion Authetication Failure traps
[2002/10/16 16:09:51] #net-snmp <TrogL> and nothing works
[2002/10/16 16:09:55] #net-snmp <TrogL> when I stop the daemons
[2002/10/16 16:10:08] #net-snmp <TrogL> some say "community mydomain"
[2002/10/16 16:10:12] #net-snmp <TrogL> some say "community public"
[2002/10/16 16:10:31] #net-snmp <TrogL> either ORACLE isn't changing domains like it should, or something isn't getting passed
[2002/10/16 16:12:51] #net-snmp <TrogL> incidentlly /var/log/snmpd.log says...
[2002/10/16 16:13:11] #net-snmp <TrogL> [init_smux] bind failed: Address already in use
[2002/10/16 16:13:35] #net-snmp <TrogL> snmd.conf line 94 error already have an entry for this process (no it doesn't, as far as I can tell)
[2002/10/16 16:20:11] #net-snmp <TrogL> line 94 is ...
[2002/10/16 16:20:13] #net-snmp <TrogL> proc mdlogd
[2002/10/16 16:21:31] #net-snmp <TrogL> its the only line with "proc" or "mdlogd"
[2002/10/16 16:22:51] #net-snmp <rstory> sorry, I'm not up to snuff on smux, and I gotta run go eat...
[2002/10/16 16:23:09] #net-snmp <TrogL> ok see you tomorrow
[2002/10/16 18:18:03] #net-snmp <gemini> HELLO!
[2002/10/16 18:18:10] #net-snmp <gemini> it's weird
[2002/10/16 18:18:43] #net-snmp <gemini> the installed net-snmp library names are "libnetsnmp"
[2002/10/16 18:19:04] #net-snmp <gemini> but why is my compiler looking for lnetsnmp ?
[2002/10/16 18:19:11] #net-snmp <gemini> and it says it can't find it
[2002/10/16 18:19:54] #net-snmp <gemini> net-snmp-config --libs shows me "lnetsnmp" is required
[2002/10/16 18:22:11] #net-snmp <gemini> if anyone is around please help me with this
--> greendisease has joined #net-snmp
[2002/10/16 20:22:16] #net-snmp <greendisease> hey guys
[2002/10/16 21:27:51] #net-snmp <gemini> hey rstory are you there
[2002/10/16 21:39:43] #net-snmp <rstory> yep...
[2002/10/16 21:40:37] #net-snmp <rstory> net-snmp-config is actually outputting the flags to add to the link step.. the flag is '-l' and netsnmp is the parameter
[2002/10/16 21:40:56] #net-snmp <rstory> it tells the linker to look for 'libnetsnmp.a' or 'libnetsnmp.so'
[2002/10/16 21:41:18] #net-snmp <rstory> what are you trying to compile?
[2002/10/16 21:48:08] #net-snmp <gemini> the demo appliaction
[2002/10/16 21:49:05] #net-snmp <gemini> # make snmpdemoapp
[2002/10/16 21:49:05] #net-snmp <gemini> cc -o snmpdemoapp snmpdemoapp.o `net-snmp-config --libs`
[2002/10/16 21:49:05] #net-snmp <gemini> ld: fatal: library -lnetsnmp: not found
[2002/10/16 21:49:05] #net-snmp <gemini> ld: fatal: File processing errors. No output written to snmpdemoapp
[2002/10/16 21:49:05] #net-snmp <gemini> make: *** [snmpdemoapp] Error 1
[2002/10/16 21:49:45] #net-snmp <gemini> but i do actually have the 'libnetsnmp.a' in my /usr/local/lib
[2002/10/16 21:51:41] #net-snmp <gemini> the flag "-lnetsnmp" tells the compiler to look for "libnetsnmp.a" ??
[2002/10/16 21:51:53] #net-snmp <rstory> yes...
[2002/10/16 21:52:12] #net-snmp <rstory> what do you get for "net-snmp-config --netsnmp-libs" ??
[2002/10/16 21:53:02] #net-snmp <gemini> hold on
[2002/10/16 21:53:05] #net-snmp <gemini> let me check
[2002/10/16 21:53:39] #net-snmp <gemini> -L/usr/local/lib -lnetsnmp
[2002/10/16 21:54:18] #net-snmp <rstory> try cc -o snmpdemoapp snmpdemoapp.o `net-snmp-config --libs` `net-snmp-config --netsnmp-libs`
[2002/10/16 21:54:39] #net-snmp <rstory> (or just add -L/usr/local/lib)
[2002/10/16 21:59:51] #net-snmp <gemini> Thanks but I still get the same error
[2002/10/16 21:59:53] #net-snmp <gemini> # cc -o snmpdemoapp snmpdemoapp.o `net-snmp-config --libs` `net-snmp-config --netsnmp-libs`
[2002/10/16 21:59:53] #net-snmp <gemini> ld: fatal: library -lnetsnmp: not found
[2002/10/16 21:59:53] #net-snmp <gemini> ld: fatal: File processing errors. No output written to snmpdemoapp
[2002/10/16 22:01:14] #net-snmp <gemini> AND my libraries are right there to show you...
[2002/10/16 22:01:14] #net-snmp <gemini> # ls /usr/local/lib
[2002/10/16 22:01:15] #net-snmp <gemini> libnetsnmp.a libnetsnmpagent.la libnetsnmpmibs.a
[2002/10/16 22:01:15] #net-snmp <gemini> libnetsnmp.la libnetsnmphelpers.a libnetsnmpmibs.la
[2002/10/16 22:01:15] #net-snmp <gemini> libnetsnmpagent.a libnetsnmphelpers.la
[2002/10/16 22:05:47] #net-snmp <gemini> HOLY SHTI!
[2002/10/16 22:05:52] #net-snmp <gemini> SORRY MAN
[2002/10/16 22:05:57] #net-snmp <gemini> MY BAD MY BAD
[2002/10/16 22:06:06] #net-snmp <rstory> aha..
[2002/10/16 22:06:15] #net-snmp <gemini> library path was not set in my profile path
[2002/10/16 22:06:17] #net-snmp <rstory> hate it when that happens
[2002/10/16 22:06:32] #net-snmp <gemini> ah sorry!
[2002/10/16 22:06:50] #net-snmp <gemini> Reall appreciate ur help though!
[2002/10/16 22:06:54] #net-snmp <rstory> np
--> jbpn has joined #net-snmp

Powered by: