Nut upsuser - Feb 2014 - Arch Linux and Tripp Lite web snmp card issues.

2014-02-25 12:58 GMT+01:00 Charles Lepple <clepple at gmail.com>:
> On Feb 24, 2014, at 11:48 PM, Jason R Begley wrote:
>
> >   0.073855     load_mib2nut: trying classic method with
'delta_ups' mib
> >   0.073878     Entering nut_snmp_get_str()
> >   0.073899     nut_snmp_get((null))
> >   0.073918     nut_snmp_walk((null))
>
> Arnaud,
>
> The gen-snmp-subdriver.sh script is setting mib2nut[i]->oid_auto_check
to
> NULL for new drivers (with no instructions on what to use in place of
> NULL). However, this seems to be only used with the "classic
method" for
> detecting an UPS.
>
> Is the intent to phase out the classic method, or should we just add an
> explicit NULL check in load_mib2nut()?
>
> The full stack trace is below:
>
> > Program received signal SIGSEGV, Segmentation fault.
> > 0xb7da9470 in __strchr_sse2_bsf () from /usr/lib/libc.so.6
> > (gdb) bt
> > #0  0xb7da9470 in __strchr_sse2_bsf () from /usr/lib/libc.so.6
> > #1  0xb7f05758 in snmp_parse_oid () from /usr/lib/libnetsnmp.so.30
> > #2  0x0804afd8 in nut_snmp_walk ()
> > #3  0x0804b261 in nut_snmp_get ()
> > #4  0x0804bab6 in nut_snmp_get_str ()
> > #5  0x0804be7b in load_mib2nut ()
> > #6  0x0804bfd2 in upsdrv_initups ()
> > #7  0x08049fb4 in main (
>
Hi Charles,

gen-snmp-subdriver.sh simply hasn't done its job, for some reasons.
well, 1 clear is that's it not a completed effort... at all
but, still IIRC, it should have produced at least a bit more

@Jason: how have you called the script?
could you please also send an archive including:
- the results of the adapted commands on #L73, #L74
https://github.com/networkupstools/nut/blob/master/scripts/subdriver/gen-snmp-subdriver.sh
- your mib file

cheers,
Arno
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.alioth.debian.org/pipermail/nut-upsuser/attachments/20140225/e9d414ac/attachment.html>

On Feb 25, 2014, at 4:23 PM, Arnaud Quette wrote:
> gen-snmp-subdriver.sh simply hasn't done its job, for some reasons.
> well, 1 clear is that's it not a completed effort... at all
> but, still IIRC, it should have produced at least a bit more
Hi Arnaud,

I may not have been specific enough in my description.

The crash that Jason saw is with the version of snmp-ups that is currently in
Git - including the new 'xppc' MIB, but the NULL pointer in question is
also in 'delta_ups'.

Given that mib2nut[i]->oid_auto_check is only used for the
"classic" OID check, I think we can just add a special case to check
for NULL, and continue the loop if oid_auto_check has not been provided.
> @Jason: how have you called the script?
> could you please also send an archive including:
> - the results of the adapted commands on #L73, #L74
>
https://github.com/networkupstools/nut/blob/master/scripts/subdriver/gen-snmp-subdriver.sh
> - your mib file

Jason,

this would still be useful to address the issue of gen-snmp-subdriver not
parsing things correctly.

-- 
Charles Lepple
clepple at gmail



-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.alioth.debian.org/pipermail/nut-upsuser/attachments/20140225/1a46c096/attachment.html>

Sorry, I just now figured out how to get a usable output.
-Option one doesn't work at all for me.

snmp_build: unknown failuresnmpget: Error building ASN.1 representation 
(Can't build OID for variable)
COUNT = 0 / 0
Creating /usr/share/snmp/mibs/ups-mib.txt-mib.h
Creating /usr/share/snmp/mibs/ups-mib.txt-mib.c
Done.

-Option 2 is attached. I ran the following to get the output.
snmpwalk -Os -v2c -c XXXXXX XXXXXXXXXXXX .1.3.6.1.4.1.850.100 -m 
//usr/share/snmp/mibs/TRIPPLITE-MIB.txt > snmpwalk-Os.log
snmpwalk -On -v2c -c XXXXXX XXXXXXXXXXXX .1.3.6.1.4.1.850.100 -m 
/usr/share/snmp/mibs/TRIPPLITE-MIB.txt > snmpwalk-On.log
./gen-snmp-subdriver.sh -s .1.3.6.1.4.1.850.100 snmpwalk-On.log 
snmpwalk-Os.log



Thanks again for the help,
Jason

On 2/25/2014 7:03 PM, Charles Lepple wrote:
> On Feb 25, 2014, at 4:23 PM, Arnaud Quette wrote:
>
>> gen-snmp-subdriver.sh simply hasn't done its job, for some reasons.
>> well, 1 clear is that's it not a completed effort... at all
>> but, still IIRC, it should have produced at least a bit more
>
> Hi Arnaud,
>
> I may not have been specific enough in my description.
>
> The crash that Jason saw is with the version of snmp-ups that is 
> currently in Git - including the new 'xppc' MIB, but the NULL
pointer
> in question is also in 'delta_ups'.
>
> Given that mib2nut[i]->oid_auto_check is only used for the
"classic"
> OID check, I think we can just add a special case to check for NULL, 
> and continue the loop if oid_auto_check has not been provided.
>
>> @Jason: how have you called the script?
>> could you please also send an archive including:
>> - the results of the adapted commands on #L73, #L74
>>
https://github.com/networkupstools/nut/blob/master/scripts/subdriver/gen-snmp-subdriver.sh
>> - your mib file
>
> Jason,
>
> this would still be useful to address the issue of gen-snmp-subdriver 
> not parsing things correctly.
>
> -- 
> Charles Lepple
> clepple at gmail
>
>
>
>
>
> _______________________________________________
> Nut-upsuser mailing list
> Nut-upsuser at lists.alioth.debian.org
> http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/nut-upsuser
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.alioth.debian.org/pipermail/nut-upsuser/attachments/20140225/5488e027/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: trippups1.tar.gz
Type: application/gzip
Size: 2493 bytes
Desc: not available
URL:
<http://lists.alioth.debian.org/pipermail/nut-upsuser/attachments/20140225/5488e027/attachment.bin>

2014-02-26 1:03 GMT+01:00 Charles Lepple <clepple at gmail.com>:
> On Feb 25, 2014, at 4:23 PM, Arnaud Quette wrote:
>
> gen-snmp-subdriver.sh simply hasn't done its job, for some reasons.
> well, 1 clear is that's it not a completed effort... at all
> but, still IIRC, it should have produced at least a bit more
>
>
> Hi Arnaud,
>
Hi Charles

> I may not have been specific enough in my description.
>
I completely overlooked that point while focusing on the new subdrv
consideration. sorry...

The crash that Jason saw is with the version of snmp-ups that is
currently
> in Git - including the new 'xppc' MIB, but the NULL pointer in
question is
> also in 'delta_ups'.
>
> Given that mib2nut[i]->oid_auto_check is only used for the
"classic" OID
> check, I think we can just add a special case to check for NULL, and
> continue the loop if oid_auto_check has not been provided.
>
you're fully right indeed!
but the root cause (and solution) is probably deeper:
https://github.com/networkupstools/nut/commit/8428fa57614ffe777191d9ae127b53d964bca6f3

however, there is still some doc completion needed for the subdriver
generation.
and lot more on this still-too-experimental code :-/

all in all, I'll have to complete quickly #20 specs:
https://github.com/networkupstools/nut/issues/20


@Jason: could you please try with the latest git master?

@Jason: how have you called the script?
> could you please also send an archive including:
> - the results of the adapted commands on #L73, #L74
>
>
https://github.com/networkupstools/nut/blob/master/scripts/subdriver/gen-snmp-subdriver.sh
> - your mib file
>
>
> Jason,
>
> this would still be useful to address the issue of gen-snmp-subdriver not
> parsing things correctly.
>
to be continued...

Arno
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.alioth.debian.org/pipermail/nut-upsuser/attachments/20140226/7b186de6/attachment.html>