Well, I'm not sure what happened this morning. Just up and quit passing
traffic and monitoring reported the slave down. My AP/master graph shows
1 registered SM during that window. Then about two hours later it just
came back up with no intervention. This is not what normally happens
though. A while after it came back up (and I woke up), I logged into the
slave and the uptime was a few hours (because it crashed last night). I
said meh and logged out. 10 minutes later it crashed and rebooted.
This is a PTP between routers running OSPF. I have a netwatch script
that sends a ping between routers which sends me an email for timeouts.
I get an email and then log into the router at the slave and see
ethernet link events in the system log (this didn't happen during the
two hours this morning). Log into the slave and the uptime is a few
minutes. So it is (hopefully was) definitely rebooting.
On 1/14/2016 11:53 AM, Josh Luthman wrote:
Put the firmware on it anyway IMO.
Is the slave side losing uptime or just losing association? Other
gear on the tower staying powered?
Josh Luthman
Office: 937-552-2340
Direct: 937-552-2343
1100 Wayne St
Suite 1337
Troy, OH 45373
On Jan 14, 2016 12:49 PM, "George Skorup" <[email protected]
<mailto:[email protected]>> wrote:
I've had this 13.7 mile link running v2.5, 5.1 band, 20MHz,
non-GPS radios for 110 days without any issues. I switched the
slave scan to include 40MHz and then moved the master to 40MHz
width the other day. Since then it appears the slave has been
rebooting. However this morning, the link wasn't passing traffic
for over two hours, then it magically fixed itself. An hour later
and it rebooted. I remember reading some stuff, possibly release
notes, that the ePTP slave could reboot under heavy interference.
This link is on a very clean channel though. About -58 RSSI and it
maintains MCS 15 both directions nearly all the time.
So I don't have a clue. Ran perfect on 20MHz channel bandwidth.
Just needed a little more throughput out of it, thus 40MHz. I just
upgraded both ends to v2.6, so I hope that fixes it.
And no, these radios are not CE+ MACs.
