Please test the v6.28 RC that we have added today. Most of the reported issues have been adressed. Let us know what issues you have and what we could still fix before release next week:
What’s new in 6.28rc5 (2015-Feb-24 09:31):
*) fixed tunnels - could crash when clamp-tcp-mss was enabled;
*) console - added ‘comment’ parameter for ‘/system script’;
*) api - return sentences can have property ~.section~ that groups values
from commands such as ~monitor~, ~traceroute~,
~print~ (with non-zero ‘interval’ value);
*) cloud - add time zone detection feature ~/system clock time-zone-autodetect~;
*) cloud - rename ~/ip cloud enabled~ to ~/ip cloud ddns-enabled~;
*) cloud - make ~/ip cloud update-time~ independent from ~/ip cloud ddns-enabled~
*) cloud - when setting ~/ip cloud ddns-enabled~ to ~no~ router will send
message to server to disable DNS name for this routerboard;
*) cloud - ~/ip cloud force-update~ command now will work also when
~/ip cloud ddns-enabled = no~. usefull if user wants to disable DDNS;
*) RB4xxGL - improved ethernet throughput (less dropped packets);
*) trafflow: - fixed crash on disable;
*) pppoe - fixed crash when big ppp packets with were sent over EOIP;
*) tilegx - fixed memory leak when queue settings are changed;
*) ar9888 - fixed crash when hw reports invalid rate;
*) console - fixed “in” operator in console;
*) console - make “/system package update print” work again.
*) console - added ‘comment’ parameter for ‘/system script’;
*) api - return sentences can have property ~.section~ that groups values
from commands such as ~monitor~, ~traceroute~,
~print~ (with non-zero ‘interval’ value);
Wasn’t that already added in 6.27? Or are there commands other than those now that also have .section?
there is very little control over the data settings as we can not set lower speeds to improve ccq and stability of the link
and we would like to see current tx power in ros
6.28rc5 killed ethernet on my test RB433AH. It’s showing some signs of life, shows in neighbours and ping sometimes works, but that’s it. Attempt to connect using WinBox (both IP and MAC) fails. And while it does, ping stops too. Then after a while it seems to get back into this not-completely-dead state. But the system is otherwise ok all the time, console works, wireless works, etc. Downgrade back to 6.27 solves it.
Update: It seems to be triggered by almost any bigger outgoing traffic and it doesn’t really have to be big in usual sense. Even an attempt to download new dlls by WinBox is enough. One sure way is to start bandwidth test in direction from router, it makes interface dead immediately (direction to router works fine).
To this ticket. I received today info from Mikrotik support that I should have all involved ros devices at the same level (6.27). But the ticket was also about the SSTP not working between two 6.27 devices. So the advice was false.
I found the solution myself: it is the PFS option that is not documented. Correct settings allows communication between both 6.27 as between different versions (for example 6.27-6.18 or 6.27-6.25 in my case).
Mikrotik should take care about the documentation if changing/adding anything!
Seeing how nobody wants anything (well, not entirely true, but this thread is pretty low volume one ), I’d like to ask one more time (there was no response from MT last time) if there’s a chance you could look at [Ticket#2014120466000688]. I need that config for few routers and the bug is blocking my ability to upgrade them. I don’t really have to do that yet, because newer versions don’t add any indispensable new feature for me, but sooner or later…
If I make it into more general question, my main concern (besides fixing the bug) is the lack of information about current status. The last response I got from MT was more than two months ago. That was after I confirmed that half of the problem was fixed. Now I can only wonder. Is someone working on the rest? Is there perhaps some problem I could help with (testing)? Did anyone look into it at all? And in case the answer is no, is it because it’s low priority, ticket was silently closed, or … ?
*) fixed tunnels - could crash when clamp-tcp-mss was enabled;
*) pppoe - fixed crash when big ppp packets with were sent over EOIP;
*) tilegx - fixed memory leak when queue settings are changed;
I have exactly these problems.
is there any ROS version where I can have fully working eoips and pppoe ?
mysz0n - How all of these three problems are manifesting themselves to you? Please write description of your problem, generate supout files which are generated after problem has appeared and send this information to our support mail. Of course please make sure that problem has not been fixed in version 6.28rc6 already.
in PPPoe Client add the option to add a column’service’ because I have several interfaces with several pppoe servers with several’service name’ so you can filter connections by’service name’
I’ve done it already [Ticket # 2015011966000455]
It turned out to be problem with eoip tunels, with the eoip enabled i got lots of kernel panic errors and reboots. (although to this day I have no answer for that ticket)
Problem with PPPoE sessions - if I’m trying to ping client, using internal network address, I have 7-8ms 0% loss if I ping him via pppoe server address I have 18-20ms and 4% packet loss.
queue - ccr suddenly begins to limit all traffic to 22Megabits on download site
I’m unable to install RC packages on working equipment.
The problem remains on ros 6.25 ad 6.27
mysz0n - Reply to ticket 2015011966000455 was sent to you already at 27/01/2015. You have not replied to it. Changes in this topic are included in version 6.28rcX. If you can not test it on your device then what makes you think that problems are still actual even in 6.28rcX? Please test it when 6.28 full release will be available and contact us if problem/s still persists.