v1.5.5 Bug Reports and Comments

[sakita]

IIRC, this was done so the switch can monitor protocols without broadcasting them... see and not be seen...

There was a user that wanted to be able to monitor protocols without announcing (broadcasting).

Unfortunately, the release notes for several versions are missing from the end of the firmware download post (which would have helped find about and when this change was made a few releases ago).

[Dave]

I will make sure Stephen provides more detailed release notes...

[Stephen]

The release note's are here:
viewtopic.php?f=17&t=5670

However lligetfa has a point I will add in the details to a tooltip in the near future.

[RebusCom]

RebusCom,

As you can see, on your log's it just say's action is "Bounce" not "Bounce Power" or "Bounce Link". I believe what has happened is that maybe the firmware you upgraded from may have only had the option to "Bounce" as I recall we separated this into Bounce Link and Bounce Power later on. If in the internal storage of the switch it still shows the action is "Bounce" then it will no longer recognize that option and nothing will happen.

To correct this, try deleting that watchdog, saving your configuration, and then re-create the watchdog and save again. This should correctly update the conditions on the switch. Then monitor the logs, if you don't see it look similar to the bolded statements in mine, aka, "Bounce Power" or "Bounce Link" then let us know here.

Upgrade was from v1.5.0 to v1.5.5. Following upgrade the watchdogs were all reflecting the desired Bounce Power, matching the previous functionality in v1.5.0. However, the behavior was not in alignment with the setting in the GUI, bouncing the link. It worked properly in v1.5.0.

I've done as you suggested toggling the setting in the GUI which will hopefully clear up the problem but I'm really surprised I'm the first to encounter this issue. There may be a lot of switches out there now with hidden watchdog misconfiguration.

[Dave]

I will have Stephen look at this for you.

[Stephen]

Upgrade was from v1.5.0 to v1.5.5

Going from 1.5.0 to 1.5.5 is a very long time to wait on firmware update's (~2 years).
It is not unheard of that unpredictable things can occur when skipping several stable release's.

Typically, the built in update utility in the switch will handle these kind's of issue's. But it has to run from the old firmware in order to upgrade to the new and sometimes when there is a large generational gap with the new firmware it fails to catch some things.

I'm really surprised I'm the first to encounter this issue. There may be a lot of switches out there now with hidden watchdog
misconfiguration.

I took a look through the code changes and found that this feature was introduced in its present form in 1.5.4, so a bit more than 5 months.
viewtopic.php?f=17&t=5670

Many people have updated since then and I have never heard of this problem before.
Assuming our troubleshooting is correct then either,
people have just been making change's regularly naturally clearing the problem,
the more incremental style of firmware update's protected their system's,
or, if the working theory is wrong, we haven't actually found the cause for your issue yet (hopefully it's not this one).

But if the steps we've taken have solved the problem and you have more switches on 1.5.0 that need to upgrade to 1.5.5 I would recommend trying to backup their configuration's after the update, and then re-apply it to the same switch. That should also clear up the issue. Using the Netonix Manager would make this process much smoother and it will let you stay up to date with the latest release's without having to click your way back to the forum's to download it the slow way.

Hopefully, if you or anyone else find's themselves in this situation, they can follow those step's to clear out this potential issue without wasting too much of their time.

[mrmarria]

Port 1 shuts down 1-8am, then comes up, a few minutes later it shuts down and is stuck down, no power draw, yet the GUI says it is on:

image

This all started happening after I tried using Watchdog to power up the port 1 if during the 1-8am period, a ping failed.
Tried bouncing manually, acts the same, reboot only way to clear apparent "on" port1




Feb 2 01:00:01 switch[1758]: Disabling PoE power on port 1 due to Shutdown Period
Feb 2 08:00:00 switch[1758]: Enabling PoE power (48VH) on port 1 due to Shutdown Period
Feb 2 08:04:55 sSMTP[990]: Creating SSL connection to host
Feb 2 08:04:56 sSMTP[990]: SSL connection using AES128-SHA
Feb 2 08:04:59 sSMTP[990]: Sent mail for GBT2@redcoyote.com (221 2.0.0 closing connection g5sm7138423ywk.46 - gsmtp) uid=0 username=Michael outbytes=85124
Feb 2 08:05:28 sSMTP[1044]: Creating SSL connection to host
Feb 2 08:05:29 sSMTP[1044]: SSL connection using AES128-SHA
Feb 2 08:05:32 sSMTP[1044]: Sent mail for GBT2@redcoyote.com (221 2.0.0 closing connection 207sm6859201ywq.100 - gsmtp) uid=0 username=Michael outbytes=85395
Feb 2 08:09:47 UI: Port 1 bounced by Michael (::ffff:192.168.77.253)
Feb 2 08:09:47 switch[1758]: Port 1 Link: changed from poe override
Feb 2 08:10:37 UI: Port 1 bounced by Michael (::ffff:192.168.77.253)
Feb 2 08:10:37 switch[1758]: Port 0 Link: re-negotiating
Feb 2 08:10:45 switch[1758]: Port 0 Link: up
Feb 2 08:10:45 switch[1758]: Port 0 Link: re-negotiating
Feb 2 08:10:53 switch[1758]: Port 0 Link: up
Feb 2 08:10:53 switch[1758]: Port 1 Link: re-negotiating
Feb 2 08:11:01 switch[1758]: Port 1 Link: up
Feb 2 08:11:22 Port: link state changed to 'up' (1G) on port 1
-snip-discard , learn , forward
Feb 2 08:11:42 Port: link state changed to 'down' on port 1
Feb 2 08:11:42 STP: msti 0 set port 1 to discarding
Feb 2 08:11:44 Port: link state changed to 'up' (1G) on port 1
-snip-discard , learn , forward
Feb 2 08:11:48 STP: MSTI0: New root on port 1, root path cost is 40020, root bridge id is 32768.B4-FB-E4-27-89-CF
Feb 2 08:13:28 sSMTP[2036]: Creating SSL connection to host
Feb 2 08:13:28 sSMTP[2036]: SSL connection using AES128-SHA
Feb 2 08:13:31 sSMTP[2036]: Sent mail for GBT2@redcoyote.com (221 2.0.0 closing connection p6sm7565039ywi.63 - gsmtp) uid=0 username=Michael outbytes=86891
Feb 2 08:14:46 Port: link state changed to 'down' on port 1
Feb 2 08:14:46 STP: msti 0 set port 1 to discarding
Feb 2 08:16:35 UI: Port 1 bounced by Michael (::ffff:192.168.77.253)
Feb 2 08:16:35 switch[1758]: Port 1 Link: changed from poe override
Feb 2 08:18:11 Port: link state changed to 'up' (1G) on port 1
-snip-discard , learn , forward
Feb 2 08:18:31 Port: link state changed to 'down' on port 1
Feb 2 08:18:31 STP: msti 0 set port 1 to discarding
Feb 2 08:18:33 Port: link state changed to 'up' (1G) on port 1
-snip-discard , learn , forward
Feb 2 08:18:38 STP: MSTI0: New root on port 1, root path cost is 40020, root bridge id is 32768.B4-FB-E4-27-89-CF
Feb 2 08:19:49 Port: link state changed to 'down' on port 1
Feb 2 08:19:49 STP: msti 0 set port 1 to discarding

[Stephen]

Please post your Power, Status, STP, and Watchdog tab. I need to replicate the behavior.

[mrmarria]

I changed the Port1 Connection With Port2 yesterday when on site, on the off chance it was that port, but the Watchdog function remains disabled until I can have another Sunday morning available to run up there on fail.

[mrmarria]

Oh small link! Done