I have a weird problem on my home RB493. I will say from the outset, I have set up dozens of Mikrotiks and I have been using them internally for a couple of years and this is a one off to me.
I have a RB2011L in the office, which I am evaluating. It is normally an RB450G. I have an IPSEC tunnel to my home where I have an RB493 with a wireless N card. I have a NAS on my home network, plugged directly into the RB493 which is the offsite replica store for the business network. My VMWare VDR (VMWare data recovery) creates a dedup backup onto a NAS in the office which then replicates to my NAS at home. The dedup store is a few metadata files and a series of "slab" files, which are the dedup chunk files, each up to 1GB in size. This makes RSYNC work quite well.
At home I run bittorrent and the usual stuff. Never have a problem. I run VOIP across the VPN to my office and from a phone to the Internet. I have a queue tree set up for QoS and mangle rules to feed it, etc. All systems work perfectly all the time.
I then initiate an rsync from the office NAS to the home NAS. Immediately, the UI on the RB493 becomes sluggish and then within a minute it crashes. I can't even get any decent log information from it before it dies, although on the odd occasion it does refresh, the CPU looks like it is going through the roof.
Any ideas? I am going to set up a syslog server to see if I can get any useful information from the logs before it crashes, but it is so hard to fault find as it almost immediately dies when I start the rsync. I have disabled the queue tree and all mangle rules to lower the load, but it makes no difference.