I relate to "crazy explanations". A lot of the 2nd rate IT crowd stays
employed by not admitting when they don't know something. Always hit the
client or boss with techno babble and they'll think you're smart,
meanwhile failure to admit that their inferences have no underlying
factual basis prevents them from actually growing. The worst part is I
think most of them actually believe their own BS. </high_horse>
If your client is using the CRS as a router then he'll be disappointed.
It does run RouterOS, but it's clearly marketed as a switch and has
clear and well advertised limits on CPU and memory. It's handy for that
one time in 20 when you need your switch to also run an IP tunnel or
provide a DHCP relay or some such
That said, yes I've had watchdog resets that go away when you disable
connection tracking. It's not necessarily RAM though because you can
actually crash a CCR 1072 with a couple gigs of traffic and connection
tracking turned on. Those have 16GB of RAM and appear to be in no
danger of running out of RAM when they die. We could only guess why.
In any case if you don't need NAT or Mangle rules then you can safely
disable connection tracking with no ill effects. As you say, that might
be worth a try on those 4011's.
On 3/10/2020 11:14 PM, Ken Hohhof wrote:
I don’t have a lot of 2011’s, but other than one that came with one
Ethernet port DOA, I don’t think I’ve seen that kind of problem on
that model. And mine are outdoors in uncontrolled environments.
Well, except for one at my office.
One just general comment, are you tracking memory utilization (like
graphing via SNMP)? I have a weirdo customer who went from using
Linux computers as routers to using Mikrotiks, I think CRS series. He
has his own /24 and gets portscanned like crazy and has connection
tracking on. He complains that his Mikrotiks keep having to be
rebooted or netinstalled and he has all sorts of crazy explanations,
but I’m convinced he is just running out of memory due to his
conntrack table filling up. Not my problem though, so I’ve avoided
getting sucked into it.
2011 has I think 128 MB RAM, the heavier duty Mikrotiks have somewhere
around 1 – 4 GB. I mean, I’ve still got a few smaller routers in my
network like 493G and 850Gx2 and I don’t see RAM running low, but just
tossing it out as a possibility.
*From:* AF <af-boun...@af.afmug.com> *On Behalf Of *ch...@wbmfg.com
*Sent:* Tuesday, March 10, 2020 6:04 PM
*To:* AnimalFarm Microwave Users Group <af@af.afmug.com>
*Subject:* Re: [AFMUG] RB4011 Lockups
Stops handing out IPs and stops passing traffic.
*From:*Colin Stanners
*Sent:*Tuesday, March 10, 2020 4:09 PM
*To:*AnimalFarm Microwave Users Group
*Subject:*Re: [AFMUG] RB4011 Lockups
What kind of issue? I haven't heard model-specific bugs with those so
they should "just work" forever.
On Tue, Mar 10, 2020, 4:57 PM <ch...@wbmfg.com
<mailto:ch...@wbmfg.com>> wrote:
We have been having a similar issue with a RB2011iL-iN .
-----Original Message-----
From: Darren Shea
Sent: Tuesday, March 10, 2020 3:31 PM
To: 'AnimalFarm Microwave Users Group'
Subject: Re: [AFMUG] RB4011 Lockups
I only have one RB4011 in the field, but it seems to have a
problem a couple
times a year. I now have it on a remote PDU which can power-cycle
it when it
stops responding to pings, but if they come out with a f/w fix for
this
problem, I'd be very interested!
-----Original Message-----
From: AF [mailto:af-boun...@af.afmug.com] On Behalf Of Nate Burke
Sent: Saturday, March 07, 2020 11:24 AM
To: Animal Farm
Subject: [AFMUG] RB4011 Lockups
This week I've had 2 RB4011 Routers lockup requiring a power
cycle. They
were still passing some traffic, but stopped doing things like handing
out DHCP, or responding to SNMP. And you couldn't log into them via
Winbox or MACNeighbor.
One was on ROS6.44.3, the other on ROS6.45. Both of them were doing
NAT, whereas most of my Mikrotiks are only routing. Both of these
locations have multiple RB4011s, and only the one doing NAT locked up.
Nothing is recorded in the log other than the reboot. These were both
routers that have been installed for months. One router lockup I can
attribute to a fluke, but 2 in the same week is worrysome. Has
any one
else noticed anything?
--
AF mailing list
AF@af.afmug.com <mailto:AF@af.afmug.com>
http://af.afmug.com/mailman/listinfo/af_af.afmug.com
--
AF mailing list
AF@af.afmug.com <mailto:AF@af.afmug.com>
http://af.afmug.com/mailman/listinfo/af_af.afmug.com
--
AF mailing list
AF@af.afmug.com <mailto:AF@af.afmug.com>
http://af.afmug.com/mailman/listinfo/af_af.afmug.com
------------------------------------------------------------------------
--
AF mailing list
AF@af.afmug.com <mailto:AF@af.afmug.com>
http://af.afmug.com/mailman/listinfo/af_af.afmug.com
--
AF mailing list
AF@af.afmug.com
http://af.afmug.com/mailman/listinfo/af_af.afmug.com
