I have problems with SwOS v2.17. From the looks of it, ICMP packets are lost.

I had to go back and downgrade to version 2.14. This version runs without these problems.

My CSS610 is connected via LACP (active) to a pfSense 2.7 (Fast mode, Layer 2,3,4) on a Supermicro X11SCL-iF board. Since the update I get messages that igb0 and igb1 interfaces are flapping at the same time.
I do not see any errors on the switch in the error tab, but sometimes errors are displayed on the link tab with impossible lengths, which after a tab change and back either don't appear anymore or appear with different values and states.

Sometimes it catches, but often dpinger exits, routing stops and internet connection hangs.

I tried it with changed settings and also with disabled IGMP snooping.

Is there any way to view more info or logs at the switch?

Can you share screenshots of the configuration on the switch and router relevant to this issue?

The screenshots are from the current running version 2.14. I didn't change anything in the configuration after the update to 2.17.
Since downgrade everything is running fine. No errors, no log entries, no interruptions.

I find it interesting that your CSS610 seems to recognise the LAG as being one and not two separate ones, despite your pfsense config showing that you use more than L2 hashing.

From https://help.mikrotik.com/docs/display/ ... Manual-LAG (emphasis by me):

IEEE 802.3ad (LACP) compatible link aggregation is supported, as well as static link aggregation to ensure failover and load balancing based only on Layer2 hashing.

Good advice. SwOS and SwOS Lite are different for LACP.
However, it does not explain why the same configuration has worked without any problems with the old version since then.
I can't find anything about LACP changes in the release notes.

I have already adjusted the LACP hash algorithm on the pfSense and change mode to passive.

When the time is right, I will attempt another update.

Hello,
I agreed, since swos lite 2.17, lag configured as active/active , active/passive or static between 2 css610 connected to the sfp+ ( Fiber cables) has strange behavior.
After a while, some devices couldn't reach anymore devices connected on the other switch.
As soon as I disabled one sfp link, all were working fine.
the same behavior happened with a css610 with 2 rj45 ports connected to 2 ports on unifi switch.
I will go back to 2.14 as soon as I can.

Two weeks ago I tried again with v2.17 and the adjusted LACP settings.
After about 2 hours the same behavior again. i.e. back to version 2.14.

Yesterday evening I upgraded to the latest version 2.18. Again I noticed the sporadic interruptions on the pfSense side, but somehow shorter.

OK, back again or trial and error?
First try: IGMP Snooping activated under System. Deactivated for all VLANs.
==> No more interruptions!!!

I don't understand this, but I am glad that it is now more stable.

At first it looked good, but after about 2 hours I had the same problems again.
After another 6 hours or so, these problems became more frequent and I had to go back to version 2.14.

It seemed to me that it had something to do with DHCP (with fallback). But even with a fixed IP, I had the same problems after a while.
The old version has now been running again for over 48 hours without an error.

Is there really no other way to view the log than via SMNP?