v1.5.10 Bug Reports and Comments

[sirhc]

FIXED/CHANGED
- Updated support for TLS v1.2 for new browser updates.

ENHANCEMENTS

KNOWN ISSUES
- DO NOT USE WITH WS-8-150-DC MODELS WAIT FOR v1.5.11 READ THREAD BELOW FOR DETAILS
- SSH KEYs broken
- POE options per port messed up for models WS-12-DC / WS-12-250-AC - NCS / WS-12-400-AC - NCS
- WEB UI issues when not at 100% Zoom on browser especially on VLAN TAB
- Some language templates need help

Released 2/16/2022

[rclark]

Thanks for a release that only addresses the browser issue!

[rclark]

Put this onto a switch to try it out, now the switch under Watts says error.

[sirhc]

Put this onto a switch to try it out, now the switch under Watts says error.

Well it would help to know the model and board rev as well as what version you upgraded from.

But sometimes on rare occasions when a switch reboots there is an error on the I2C bus. This is not new to v1.5.10 but has existed for ever, itis just the nature of the I2C channel. It happens so rarely that we never looked to much into it as to fix something you need to be able to recreate it. Kinda like taking your car to a mechanic and saying well it was making this noise but its not now can you fix it. It more often occurred on DC models usually when not hooked directly to battery bank but to charging controller bypass or output terminals.

Sometimes a soft reboot fixes it sometimes a power cycle is required.

[rclark]

Upgraded from 1.5.8, switch is a WS-8-150-DC. Reboot did nothing. This switch is on a mountain in a remote location, it will be quite a while before we go there.

I see there are some i2c utilities on the switch, can one or more of those be used to help?

[sirhc]

Upgraded from 1.5.8, switch is a WS-8-150-DC. Reboot did nothing. This switch is on a mountain in a remote location, it will be quite a while before we go there.

It sounds like an issue I recently (2 months ago) worked with a guy named Chris in Jamaica with, he also had a WS-8-150-DC off grid on a mountain that lost power due to clouds for many days and depleted his batteries and he had the sleep and wake up voltages too close and insufficient batteries so it was going to sleep and waking up in a cycle to it was obviously rebooting every so many minutes and at one point there was an I2C error which caused the same issue and remote rebooting would not correct it. The following day his crew went out fully charged the batteries and did a power off for 10 seconds then on and the error cleared. They also spread the hibernate and wake up voltages and added another battery and the problem has not returned.

The good news is if the unit is up and just not reporting watts it should function until you get out to it.

[sirhc]

I just looked over my notes for testing of v1.5.10 yesterday and today.

Don't panic but WS-8-150-DC happens to be one model I did not test but I highly doubt this is related to that as all we had to do with v1.5.10 is replace a module for TLS v1.2 and did not even have to replace lighthttpd since we did not go to TLS v1.3 which really was no benefit for our application.

No other code had to be changed between v1.5.9 and v1.5.10 and I did test v1.5.9 with WS-8-150-DC and from version v1.58 to v1.5.9 also had no changes that would effect these functions so I am pretty confident that this is one of those rare occurrences when the I2C bus gets confused and requires a power cycle to allow the board to completely discharge as a soft reboot does not power down the board which is why with many electronics not just our switches sometimes need a power cycle to fully discharge everything to clear the error.

[rclark]

I downgraded to 1.5.8 and the error was gone. I then put 1.5.10 back on and the error returned also noticed in the upper right for input current it says NaN.undefined A. Put 1.5.8 back on and the errors are gone.

[sirhc]

I downgraded to 1.5.8 and the error was gone. I then put 1.5.10 back on and the error returned also noticed in the upper right for input current it says NaN.undefined A. Put 1.5.8 back on and the errors are gone.

Hmmm. Interesting.

I will dig out a WS-8-150-DC tomorrow and test it.
Let me look over my notes here and report all models I did test on here in a few minutes

[sirhc]

OK here are the models I tested v1.5.10 on and worked fine. Also note I did "most" upgrades from v1.5.9rc1 and v1.5.9 and 1 or 2 from v1.5.8

THE FOLLOWING MODELS ALL TESTED FINE
WS-6-MINI
WS-6-MINI - NCS
WS-8-150-AC
WS-12-250-AC
WS-12-250-AC - NCS
WS-12-250-DC
WS-12-250-DC - NCS
WS-12-400-AC
WS-26-400-AC
WS-26-500-DC
WS-26-400-IDC
WS-26-400-IDC - NCS

THE FOLLOWING MODELS ALL TESTED FINE - END OF LIFE MODELS - OLD
WS-8-250-AC
WS-12-250A
WS-24-400A

NOT TESTED BUT REALLY SHOULD NOT MATTER EXCEPT WS-8-150-DC
WS-8-150-DC - DO NOT USE WITH THIS MODEL WAIT FOR v1.5.11 - DOES NOT WORK PROPERLY
WS-8-150-DC - NCS - DO NOT USE WITH THIS MODEL WAIT FOR v1.5.11- DOES NOT WORK PROPERLY
WS-10-250-AC - END OF LIFE - OLD - SHOULD BE OK
WS-26-500-DC - NCS - SHOULD BE OK

WS-24-400B - END OF LIFE OLD - SHOULD BE OK

Out of all the models that I guess "might" "possibly" have an issues would be the WS-8-150-AC/DC models because I "think" there is something about those models with the addresses on A2D at the MUX from the SFP ports on the DC model?