-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 After digging through snmpwalks and gets, I may have found what was causing problems with voltages. I'll update more later.
More important is the utter failure of wireless in the 2.4 with 10mhz channels. On the one AP that we tried, which had a mix of Ubiquiti and Tranzeo clients all reliability went out the window. HAVE to run noise suppression to even get it to work. Not necessary with 4.11.... This could be the end of our use of MT for wireless. Robert On 09/22/2010 08:47 AM, Randy Cosby wrote: > One other thing I noticed upgrading some units to 5.0RC1 - make sure > you have the firmware up to date (/system routerboard print then /system > routerboard upgrade) before you try upgrading. If you are more than 1 > revision behind 2.27, update to 4.11 or another 5.0 beta before going to > 5.0RC1. I lost ipconfig and some vlan config on one that jumped too > many revisions, and on another I had to reload from netinstall. Luckily > it was in my office, not on a mountaintop. > > > On 9/22/2010 9:34 AM, Bill Prince wrote: >> I plugged in a unit running 5.0RC1 yesterday, and added it to our >> cacti device list. The stock cacti template causes something to >> happen during an poll or cacti query to cause the unit to go to 100% >> CPU busy, and to partially lock up. >> >> However, a regular snmpwalk, or snmpget does not seem to do it. >> >> I had to reboot the unit to get it out of that state. >> >> If I declare the device as "generic" instead of "mikrotik" it seems to >> work ok. Problem is, there are something like 80 OIDs probed in the >> stock template. >> >> Did not send in a bug report, because I'm not sure what's going on. >> >> bp >> >> >> On 9/20/2010 10:40 PM, Butch Evans wrote: >>> On Mon, 2010-09-20 at 18:08 -0600, Travis Johnson wrote: >>>> I thought nv2 (and airmax) were both using the hardware timing on the >>>> wireless card? That should reduce CPU compared with previous versions. >>> I am only reporting my tested results. The timing is, as you state, >>> done in hardware for nv2. The framing, however, is done on the host >>> processor (for both nv2 and nstreme). I have not finished testing the >>> nv2 protocol and may have had some inefficient settings, so this test >>> may change. We will be covering nv2 and testing nv2 a little more >>> extensively next week at the training. >>> >> _______________________________________________ >> Mikrotik mailing list >> Mikrotik@mail.butchevans.com >> http://www.butchevans.com/mailman/listinfo/mikrotik >> >> Visit http://blog.butchevans.com/ for tutorials related to Mikrotik >> RouterOS > -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.10 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org/ iQEcBAEBAgAGBQJMmluqAAoJEC+8HUjSuDOsCOkH/RkLpwKzitc2a4QdQfIuRPMF Cu/PRS5Zz5FzciU22/qa+7tmBZC7VYdsdAVSw0VKp2RYrTHNDzmNf3weux3v1x/9 r8NREaaoOBWI5ybu1P1Dn33Q6g/Y25mwogfJUCr9ygu0C2bP4imceOfPLvFiKc6l 5FKViLPTfSh+CPZg7hP5ml6CqwBXKmEP0lrK7ZG8yhkGvJezUqhA7uWiQJ59jF90 hgpEsiC1IIX6URYf1hiCtWNi5A0D1kTPPHaDGXPKm6R5Cm+uZonOoffRfbUs2aWB DVJCgo9cDDXBosPfqh50kAvxYBBbQ9QT6AjQrjBiazGEjpEvppnEQF/4MevfGn8= =WkJb -----END PGP SIGNATURE----- _______________________________________________ Mikrotik mailing list Mikrotik@mail.butchevans.com http://www.butchevans.com/mailman/listinfo/mikrotik Visit http://blog.butchevans.com/ for tutorials related to Mikrotik RouterOS
