Netonix Forums

Please post comments and reports bugs with v1.3.7 FINAL in this thread

v1.3.7 wrote:
FIXES
- Fixed issue with RADIUS passwords that have special characters
- Fixed flow control on quad SFP ports

ENHANCEMENTS
- Added auto negotiation to quad SFP ports
- Added more meters to the Device Status page

KNOWN ISSUES
- Last know IP address on MAC table introduced in v1.3.0 causes high CPU usage if running a large FLAT network
- Ubiquiti Discovery Protocol Still Broken
- IPERF is still a work in progress
- After upgrading to v1.3.7 make sure to read this POST
viewtopic.php?f=17&t=1180&p=8865#p8865

Released 11/6/2015

currently getting a 404 trying to download 1.3.7...

Following up on the i2c errors from the 1.3.6 thread. Have a WS-12-250-AC Rev D - running 1.3.7rc4 and started seeing errors for power and temp, plus errors in the logs (below). I tried upgrading to 1.3.7rc7 and same thing. 1.3.7 final appears to be missing from the download folder.

I saw a full power cycle is recommended, but won't be able to do that for at least a few hours. Anything I can grab from the command line that would help in the mean time?

Code: Select all

Nov 10 06:08:49 switch[1421]: i2c_read failed after 5 tries: addr 49: Failed writing command to slave
Nov 10 06:08:52 switch[1421]: i2c_read failed after 5 tries: addr 49: Failed writing command to slave
Nov 10 06:08:52 switch[1421]: Failed to get board temp
Nov 10 06:08:53 switch[1421]: Error reading temp sensor
Nov 10 06:08:55 switch[1462]: Detected warm boot
Nov 10 06:09:03 switch[1421]: retrying i2c
Nov 10 06:09:04 switch[1421]: i2c error: excessive arbitration lost start errors (0x51)
Nov 10 06:09:15 switch[1421]: retrying i2c
Nov 10 06:09:15 switch[1421]: i2c error: excessive arbitration lost start errors (0x51)
Nov 10 06:09:26 switch[1421]: retrying i2c
Nov 10 06:09:26 switch[1421]: i2c error: excessive arbitration lost start errors (0x51)
Nov 10 06:09:37 switch[1421]: retrying i2c
Nov 10 06:09:40 switch[1421]: i2c_read failed after 5 tries: addr 49: Failed writing command to slave
Nov 10 06:09:43 switch[1421]: i2c_read failed after 5 tries: addr 49: Failed writing command to slave
Nov 10 06:09:43 switch[1421]: i2c error: excessive arbitration lost start errors (0x51)
Nov 10 06:09:54 switch[1421]: retrying i2c
Nov 10 06:09:57 switch[1421]: i2c_read failed after 5 tries: addr 49: Failed writing command to slave
Nov 10 06:10:00 switch[1421]: i2c_read failed after 5 tries: addr 49: Failed writing command to slave


nickwhite wrote:I saw a full power cycle is recommended...

I saw i2c errors on 1.3.6 two hours after flashing a WS-6-MINI. It is powered from a 48VH port of a WS-24 so I did a port bounce but that did not resolve it.

I swapped out the MINI and tried to repro the i2c on the bench but could not. I put it back in service and it has been well behaved. All Chris and I can think of, is that the short 5 second power cut of the port bounce was not long enough to clear out some bad bits in RAM.

Chris put up 1.3.7 last night and then promptly took it down. Not 100% sure of the reason. He said he will put it back up today after conferring with Eric.

Hi Nick,

v1.3.7 will be out soon but normally when this happens (which is very rarely) a power cycle is often needed to clear it.

I would wait until I put v1.3.7 up then upgrade it and if needed try a power cycle where you unplug it for at least 10 seconds.

You can now download v1.3.7, sorry for the delay.

Loaded on my DC12, looking good


Minor bug noted ...
Defaulted switch, everything fine and good.
Setup IP, device name, and password
Applied but did not catch the network change in time, so the switch reverted, except the password did not , no biggy I knew what it was.

I have ports 2,3, and 4 configured for 48VH, 24VH, 24VH . I have no radios plugged in yet, I am seeing ports 2 and 4 reporting wattage that fluctuates between 0 and 1.4 on port2 and between 0 and 0.8 on port4 , weird.

DC12 switch BTW

sirhc wrote:Hi Nick,

v1.3.7 will be out soon but normally when this happens (which is very rarely) a power cycle is often needed to clear it.

I would wait until I put v1.3.7 up then upgrade it and if needed try a power cycle where you unplug it for at least 10 seconds.

Thanks. Power cycle seemed to fix the i2c sensor issue. I'll upgrade to 1.3.7 tonight.

rockhead wrote:I have ports 2,3, and 4 configured for 48VH, 24VH, 24VH . I have no radios plugged in yet, I am seeing ports 2 and 4 reporting wattage that fluctuates between 0 and 1.4 on port2 and between 0 and 0.8 on port4 , weird.

DC12 switch BTW

OH, you are seeing "phantom" watts meaning you see 0 then this weird .6 or something comes in for a second.

Damaged boards show a constant watts

Our current sensors are not real accurate at extremely low watts. We try to filter these phantom watts out but I it is pretty hard. It is a trade off for accuracy on a real load or an extremely "low" load