Firmware v1.1.7 bug reports and comments

[sirhc]

Please post comments and bus for v1.1.7 in this thread.

FIXES
- Fix where the mac used as the basis for the ipv6 local link address was not being properly set.
- Additional fixes for Firefox and Safari - Feel the love Firefox/Safari users

KNOWN BUGS
- If using DHCP and you change the Default/Management VLAN ID after the switch gets an address is does not get a new address.
- Not all CLI commands are present.
- In the CLI some voltages are reported incorrectly with the "sh status" command
- In the CLI not all command are present to configure VLANs

[lligetfa]

NTP failed to set the date/time after flashing. I had to disable NTP and then enable it.

[NickOlsen]

After the update I had to reset my timezone to America/Newyork. But the time under status was always right. Only the log times were off.

[cbl]

I just upgraded a brand new WS-12-250A we got today from 1.1.0rc19 to 1.1.7. It ended up needing a power cycle to be reachable again. I waited about 4-5 minutes after pings stopped post-upgrade to see if it'd come up. It had zero config on it, just immediately tried upgrading it out of the box on my desk.

[sirhc]

I just upgraded a brand new WS-12-250A we got today from 1.1.0rc19 to 1.1.7. It ended up needing a power cycle to be reachable again. I waited about 4-5 minutes after pings stopped post-upgrade to see if it'd come up. It had zero config on it, just immediately tried upgrading it out of the box on my desk.

We have seen this reported by several users that after upgrading the switch needed rebooted. We are testing a fix for this now and I hope to post a new firmware version to correct this issue.

It has been hard to track down as it happens so rarely, for instance I have never seen it on any of my towers at my wisp and I upgrade them all the time. I have been able to recreate it in a LAB but it is not easy and requires numerous reboots and or firmware upgrades before it happens.

We think we found it and we are testing it now but when you are talking about something that occurs only 1% of the time it is hard to recreate but since we now have well over 1,000 switches in the field 1% shows up every time we push a new firmware as 1% of 1,000 is 10 occurrences.

The good news is we "think" we found the cause, I am testing now but as I said it is hard to recreate on an individual basis.