--> Gromit has joined #net-snmp
[2003/01/28 02:06:08] #net-snmp <Gromit> hoi ...
[2003/01/28 02:06:58] #net-snmp <Gromit> I have some issues with snmp for checking some cisco hardware ...
[2003/01/28 02:07:38] #net-snmp <Gromit> I do an snmpwalk and see OIDs for all interfaces on the cisco but when I do a snmpget on an interface I will check,
[2003/01/28 02:08:20] #net-snmp <Gromit> I get something like Error in packet
[2003/01/28 02:08:21] #net-snmp <Gromit> Reason: (noSuchName) There is no such variable name in this MIB.
[2003/01/28 02:08:21] #net-snmp <Gromit> Failed object: interfaces.ifTable.ifEntry.ifAdminStatus.50
[2003/01/28 02:08:36] #net-snmp <Gromit> what is going wrong here?
--> Jo-Con-El has joined #net-snmp
[2003/01/28 03:05:18] #net-snmp <Jo-Con-El> Hello!
[2003/01/28 03:05:27] #net-snmp <Jo-Con-El> I have a problem with snmptrapd
[2003/01/28 03:05:33] #net-snmp <Jo-Con-El> It's going crazy...
[2003/01/28 03:07:28] #net-snmp <Jo-Con-El> Hello!
--> dts has joined #net-snmp
[2003/01/28 03:56:09] #net-snmp <potifer> Don't you people have better things to do at 4 in the morning?
[2003/01/28 04:07:22] #net-snmp <dts> Yes - but it's just gone noon here in the UK.
[2003/01/28 04:19:33] #net-snmp <Jo-Con-El> potifer: Are you talking 'bout me?
--> wes has joined #net-snmp
[2003/01/28 06:39:17] #net-snmp <rstory> Gromit: do you seee ifAdminStatus.50 when you do a snmpwalk?
[2003/01/28 06:53:19] #net-snmp <Gromit> rstory: yes ..
[2003/01/28 07:00:36] #net-snmp <Gromit> rstory: I get this with snmpwalk
[2003/01/28 07:00:36] #net-snmp <Gromit> interfaces.ifTable.ifEntry.ifAdminStatus.50 = up(1)
--> Entity_ has joined #net-snmp
[2003/01/28 07:02:27] #net-snmp <Entity_> Is hrStorageSize.1, and hrStorageUsed.1 an unreiable means of determing disk space
[2003/01/28 07:02:29] #net-snmp <Entity_> ?
[2003/01/28 07:03:06] #net-snmp <wes> should be ok, I think. What's the problem?
[2003/01/28 07:03:22] #net-snmp <Entity_> returns a different value all time
[2003/01/28 07:03:52] #net-snmp <Entity_> mrtg sadi my disk was 27 gigs last night today it said 700 gigs and now i get nothing
[2003/01/28 07:05:36] #net-snmp <rstory> Gromit: but a snmpget of the same object fails?
[2003/01/28 07:05:37] #net-snmp <Entity_> perhapes I should monitor from dskTable instead and monitor partisions
[2003/01/28 07:05:50] #net-snmp <Gromit> rstory: right ...
[2003/01/28 07:06:13] #net-snmp <Gromit> rstory: I am pretty new to snmp .. and I not know, where to look, whats wrong here ...
[2003/01/28 07:06:36] #net-snmp <Gromit> rstory: I included the cisco-MIBs as well ..
[2003/01/28 07:06:55] #net-snmp <rstory> Gromit: it sounds like a problem with the cisco hardware... what is the exact snmpget command you are using?
[2003/01/28 07:07:10] #net-snmp <Entity_> ok well I think I'm gonna try with dsktable.. s how would I compile it in ?
[2003/01/28 07:07:12] #net-snmp <Entity_> UCD-SNMP-MIB.txt
[2003/01/28 07:07:15] #net-snmp <Entity_> I need that
[2003/01/28 07:07:32] #net-snmp <Gromit> rstory: mompl
[2003/01/28 07:07:55] #net-snmp <wes> Entity_: it's already compiled in. YOu just need to turn it on. Put "disk /something" in your snmpd.conf file.
[2003/01/28 07:08:09] #net-snmp <wes> what architecture are you using this on?
[2003/01/28 07:08:14] #net-snmp <Entity_> linux
[2003/01/28 07:08:18] #net-snmp <Entity_> and solaris
[2003/01/28 07:08:21] #net-snmp <Entity_> linux for now
[2003/01/28 07:08:43] #net-snmp <wes> hmmm... the disk size should definitely be right under linux.
[2003/01/28 07:08:51] #net-snmp <Gromit> sudo /usr/bin/snmpget -m ALL -v 1 -c $COMUNITY $HOST .1.3.6.1.2.1.2.2.1.7.50
[2003/01/28 07:08:52] #net-snmp <wes> sounds like a mrtg configuration problem.
[2003/01/28 07:09:14] #net-snmp <Entity_> how could I test snmp ?
[2003/01/28 07:10:53] #net-snmp <wes> use snmpwalk to walk the table and see if the values are right.
[2003/01/28 07:11:21] #net-snmp <Entity_> stuff jus pours past so fast and doesn't stp when I do that
[2003/01/28 07:11:33] #net-snmp <Entity_> whats weird is.. this ws workin last night
[2003/01/28 07:11:44] #net-snmp <Entity_> I have made no changes infact I didn't even close my terminal
[2003/01/28 07:12:40] #net-snmp <Entity_> Now I get no errors and nothing on my chart
--> miike has joined #net-snmp
[2003/01/28 07:13:27] #net-snmp <Gromit> rstory: any hints besides, that there would be a problem with the cisco?
[2003/01/28 07:13:41] #net-snmp <Gromit> rstory: it is a cisco 7206vxr
[2003/01/28 07:13:41] #net-snmp <Entity_> And it scares me to think I gotta set this up on 30 systems
[2003/01/28 07:14:34] #net-snmp <wes> well, I'm not really sure exactly what the problem might be.
[2003/01/28 07:14:50] #net-snmp <wes> you could also run "snmpdf" on the host and see what it reports your disks look like.
[2003/01/28 07:15:09] #net-snmp <wes> If that looks about right then it's likely an mrtg problem.
[2003/01/28 07:15:35] #net-snmp <Entity_> do ya know the sytax for that command
[2003/01/28 07:15:39] #net-snmp <Entity_> syntax
[2003/01/28 07:16:28] #net-snmp <Entity_> ok do you knwo something better for what I need then mrtg ?
[2003/01/28 07:16:56] #net-snmp <Entity_> I need to monitor cpu/desk/mem across 20 linux and 10 solaris boxes
[2003/01/28 07:17:03] #net-snmp <Entity_> disk
[2003/01/28 07:17:42] #net-snmp <wes> mrtg probably isn't the problem, it's just a configuration issue.
[2003/01/28 07:17:48] #net-snmp <rstory> Gromit: do other snmpget commands work? for example, sysLocation.0 (instead of .1.3.6...)
[2003/01/28 07:17:49] #net-snmp <Entity_> and graph it
[2003/01/28 07:17:58] #net-snmp <wes> I haven't tried to do what you are doing so I probably can't help. mrtg should be just fine to graph it though.
[2003/01/28 07:18:09] #net-snmp <Entity_> http://www.net-snmp.org/tutorial-5/mrtg/index.html
[2003/01/28 07:18:09] #net-snmp <wes> snmpdf -v 1 -c COMMUNITY localhost
[2003/01/28 07:18:13] #net-snmp <Entity_> thats what I am useing
[2003/01/28 07:18:20] #net-snmp <wes> that's the right place to start.
[2003/01/28 07:18:50] #net-snmp <Gromit> rstory: yes ..
[2003/01/28 07:18:51] #net-snmp <Gromit> jpa@beverly:~$ sudo /usr/bin/snmpget -m ALL -v 1 -c gZLBFiKC kiel1.bb.addix.net sysLocation.0
[2003/01/28 07:18:55] #net-snmp <Gromit> system.sysLocation.0 =
[2003/01/28 07:19:06] #net-snmp <Gromit> and with snmpwalk I get system.sysLocation.0 =
[2003/01/28 07:19:07] #net-snmp <Entity_> ok wes that looks fine
[2003/01/28 07:19:23] #net-snmp <Gromit> rstory: it only with parts of the interfaces ... I can see ..
[2003/01/28 07:21:16] #net-snmp <wes> It might be wise to simply delete your current mrtg data and see if it returns to normal (if it worked yesterday till you left work....)
[2003/01/28 07:22:00] #net-snmp <Entity_> well .. it wasn't correct yesteray either it was 7 gigs too large
[2003/01/28 07:25:25] #net-snmp <rstory> Gromit: first, i hope the isn't your real community string for a real box on the net. if you, you might want to consider changing it.
[2003/01/28 07:25:47] #net-snmp <rstory> Gromit: i think you are going to have to have to talk to cisco.. they have a bug... eithe (a) they don't handle snmpget correctly or (b) the box is configured to prevent access to the object, but a bug is letting you see it on a walk
[2003/01/28 07:27:07] #net-snmp <rstory> what do you get if you try snmpgetnext [parms] ifAdminStatus.49 ?
[2003/01/28 07:27:13] #net-snmp <Entity_> I'm blowin it all away and resetting everything
[2003/01/28 07:27:45] #net-snmp <Gromit> rstory: the same as .50
[2003/01/28 07:28:01] #net-snmp <Gromit> I will check .. when I get a status when I get an error
[2003/01/28 07:29:04] #net-snmp <rstory> try snmpgetnext with whichever index is the one snmpwalk shows previous to 50
[2003/01/28 07:30:05] #net-snmp <rstory> also not sure why you are using sudo for snmpget, only the daemons (snmpd, snmptrapd) need to be run as root..
[2003/01/28 07:32:11] #net-snmp <Gromit> rstory: it works up to 42 ... all above gives mir errors back
[2003/01/28 07:32:55] #net-snmp <rstory> if some work, and some don't, that sounds more and more like it's a cisco problem
[2003/01/28 07:33:31] #net-snmp <Gromit> yeah ... I think so ..
[2003/01/28 07:34:18] #net-snmp <Gromit> rstory: on an other cisco, a 7206vxr too, I got for one Interface ifstatus (up) and for operstatus (not present)
[2003/01/28 07:34:24] #net-snmp <Gromit> rstory: what can this be?
[2003/01/28 07:40:59] #net-snmp <rstory> i guess they don't implement operstatus
[2003/01/28 07:41:37] #net-snmp <rstory> who is telling you not present? that doesn't sound like one of our messages
[2003/01/28 07:43:48] #net-snmp <Gromit> rstory: I got this with snmpwalk and snmpget
--> TrogL_AFK has joined #net-snmp
[2003/01/28 10:15:21] #net-snmp <Entity_> ok it's just not gonna work
[2003/01/28 10:27:58] #net-snmp <Entity_> ok I cannot get mrtg to pull any info from net-snmp
[2003/01/28 10:28:44] #net-snmp <Entity_> wheather it be monitoring interfaces.. disk,cpu .. it doesn't matter .. yet snmpwalk is able to get the info
[2003/01/28 10:40:37] #net-snmp <wes> sounds like an access control problem.
[2003/01/28 10:46:10] #net-snmp <Entity_> well .. if you don't have the comunity set right .. mrtg will just say " no Answer "
[2003/01/28 10:46:54] #net-snmp <Entity_> what hapens is the graph comes up but with no data on it
[2003/01/28 10:47:44] #net-snmp <wes> odd.
[2003/01/28 10:47:56] #net-snmp <wes> tcpdump might be your friend (or better yet: ethereal)
[2003/01/28 10:48:56] #net-snmp <Entity_> funny .. I use snmpget and it shows the info no prob
[2003/01/28 10:49:20] #net-snmp <Entity_> snmpget -m ALL -v 1 -c public 172.18.1.11 hrStorageUsed.1
[2003/01/28 10:49:32] #net-snmp <Entity_> HOST-RESOURCES-MIB::hrStorageUsed.1 = INTEGER: 343208
[2003/01/28 10:49:36] #net-snmp <Entity_> come sback fine
[2003/01/28 10:49:55] #net-snmp <Entity_> now id it will just take that number and grahp it
[2003/01/28 10:53:21] #net-snmp <Entity_> the problem is I don't know if it is really asking and if there is really an answer
[2003/01/28 11:04:58] #net-snmp <Entity_> they are talking
[2003/01/28 11:05:13] #net-snmp <Entity_> but don't se the request for anything other in uptime
[2003/01/28 11:05:24] #net-snmp <Entity_> other than
[2003/01/28 11:07:22] #net-snmp <Entity_> 14:02:24.810267 arp who-has pc-jduhault.tor.vixs tell fs001.tor.vixs
[2003/01/28 11:07:22] #net-snmp <Entity_> 14:03:18.343929 netmon.tor.vixs.32789 > fs001.tor.vixs.snmp: GetRequest(39) 25.2.3.1.5.1 .iso.org.dod.internet.mgmt[|snmp] (DF)
[2003/01/28 11:07:22] #net-snmp <Entity_> 14:03:18.344966 fs001.tor.vixs.snmp > netmon.tor.vixs.32789: GetResponse(39) 25.2.3.1.5.1=1033725 .iso.org=[|snmp] (DF)
[2003/01/28 11:07:25] #net-snmp <Entity_> I see that
[2003/01/28 11:07:54] #net-snmp <Entity_> forget the first line
--> benr has joined #net-snmp
[2003/01/28 11:54:01] #net-snmp <TrogL> wes, robert...I've got a FAQ and README to go with those binaries
[2003/01/28 11:54:39] #net-snmp <TrogL> how do you want it?
[2003/01/28 12:43:38] #net-snmp <TrogL> does linux use ELF binaries??
[2003/01/28 13:49:49] #net-snmp <benr> ACTION is away: they says I have to work.... or like they won't pay me.
[2003/01/28 14:01:53] #net-snmp <TrogL> ACTION nudges Robert
[2003/01/28 14:09:34] #net-snmp <rstory> ACTION looks around, confused
[2003/01/28 14:10:21] #net-snmp <TrogL> "wes, robert...I've got a FAQ and README to go with those binaries"
[2003/01/28 14:10:29] #net-snmp <TrogL> "how do you want it?"
[2003/01/28 14:12:16] #net-snmp <rstory> i'd like mine in chocolate letters on a nice yellow cake...
[2003/01/28 14:12:28] #net-snmp <rstory> but seriously, just email it to us
[2003/01/28 14:35:33] #net-snmp <benr> ACTION is back (gone 00:45:44)
--> TrogL has joined #net-snmp
--> miike has joined #net-snmp
--> Entity_ has joined #net-snmp
--> wes has joined #net-snmp
--> rstory has joined #net-snmp
--> potifer-tempcoma has joined #net-snmp
--> ranc0r has joined #net-snmp
[2003/01/28 17:27:46] #net-snmp <ranc0r> hello. i'm having a strange situation where if I snmpwalk an OID I get a certain value but if i try to snmpget that same OID it returns 0. Has anyone ever seen something like that?
[2003/01/28 17:28:37] #net-snmp <rstory> hmm.. no, but someone earlier had a similar problem, except the get failed completely.. would you by change be talking to a cisco box?
[2003/01/28 17:29:03] #net-snmp <ranc0r> no its a wireless bridge/router made by a company called solectek
[2003/01/28 17:29:37] #net-snmp <rstory> hmm... actually, you should expect different results from a walk and get of *exactly* the same oid..
[2003/01/28 17:29:53] #net-snmp <rstory> or do you mean you take an OID you see during snmpwalk, and use it for a get?
[2003/01/28 17:30:10] #net-snmp <ranc0r> the OID i see during snmpwalk
[2003/01/28 17:30:42] #net-snmp <ranc0r> enterprises.2890.1.1.31.1.1.14.256 = "30" vs. enterprises.2890.1.1.31.1.1.14.256 = "0"
[2003/01/28 17:31:30] #net-snmp <rstory> what about a snmpgetnext enterprises.2890.blah.blah.255 ?
[2003/01/28 17:32:30] #net-snmp <ranc0r> i've installed the mib, but it doesn't seem to like using text, i have to use the numbers
[2003/01/28 17:32:54] #net-snmp <rstory> snmpwalk -m ALL ....
[2003/01/28 17:33:41] #net-snmp <ranc0r> heh. enterprises.solectek.solproducts.sw.31.1.1.14.256 = "28"
[2003/01/28 17:34:16] #net-snmp <rstory> what if you do the getnext a fews times in a row, quickly?
[2003/01/28 17:34:48] #net-snmp <rstory> (and try the same with a basic snmpget too)
[2003/01/28 17:35:30] #net-snmp <ranc0r> hm getnext returns the right value
[2003/01/28 17:36:19] #net-snmp <rstory> well, it wouldn't be the first time someone had a broken agent implementation...
[2003/01/28 17:36:20] #net-snmp <ranc0r> snmpget never does though. i just han it 15 or so times
[2003/01/28 17:37:04] #net-snmp <ranc0r> UCD-snmp version: 4.2.5 <-- would upgrading help?
[2003/01/28 17:37:14] #net-snmp <ranc0r> debian seems to be a bit behind
[2003/01/28 17:37:19] #net-snmp <rstory> i very seriously doubt it...
[2003/01/28 17:37:49] #net-snmp <rstory> 4.2.6 is the latest version of the old-api code, so they aren't too far behind...
[2003/01/28 17:37:56] #net-snmp <ranc0r> could it be a mib problem? I talked with the company and the say they've only used it with SNMPc
[2003/01/28 17:38:15] #net-snmp <rstory> i can see why they'd wait to move to the 5.0.x branch
[2003/01/28 17:38:31] #net-snmp <rstory> no, the mib is on your machine, and has no effect on the data the agent send back
[2003/01/28 17:38:53] #net-snmp <ranc0r> oh i see,
[2003/01/28 17:40:31] #net-snmp <ranc0r> i'll just edit my script to use snmpgetnext for that oid. maybe i'll hit the company with a bug report.
[2003/01/28 17:40:46] #net-snmp <ranc0r> thanks for your help.