EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: sequoia on September 27, 2015, 06:22:31 am
-
I just got new Rigol, and noticed network related issues with it. Are these known issues or is this possibly bad unit (otherwise unit seems to work perfectly, though)?
1) If trying to access the web interface of the unit (http://x.x.x.x/ (http://x.x.x.x/)), web interface and other ports stop responding. Sometimes I get the status page only partially (and network side is frozen) and sometimes it loads ok, but freeze happens when trying to re-load it or access any other pages. Interesting thing is that unit stops completely responding on all (TCP) ports when this happens. Only ping works after the freeze. Only restarting the scope seems to make it recover. However, the network stack "crash" doesn't seem to affect operation of the scope itself.
Accessing port 5555 seems to work fine (doesn't trigger this issue)....
2) DHCP client is very buggy. I observed unit requesting a IP/lease normally when its powered on, but it keeps then renewing the lease every 2-3 minutes, completely ignoring the actual lease time (in my case 24hours or 86400 seconds). Sane behaviour would be to wait at least about half the lease time before checking again with the DHCP server...
I have the latest firmware 00.04.03.SP1 (aka 00.04.03.01.05)
-
I have just tested - Mine is doing exactly the same. Freezes up, when trying to access other pages, etc. The scope UI is responsive though.
-
Yep, same here.
Fortunately, the webinterface is useless anyway so nobody cares...
-
I guess the "good" news is that this seems like firmware bug and not that my new scope has hardware issue.
But, this points to pretty much non-existent QA process at Rigol if this kind of issues get through to released firmware.
At minimum this is a DoS vulnerability, but on the other hand it might indicate buffer overflow in the unit that might allow clever exploit to be used to gain "root" access to the device...
Another note, there seems to be lot of ports open on these things (and even ancient Portmap/RPCbind service):
PORT STATE SERVICE
80/tcp open http
111/tcp open rpcbind
617/tcp open sco-dtmgr
618/tcp open dei-icda
619/tcp open compaq-evm
5555/tcp open freeciv
5566/tcp open westec-connect
6666/tcp open irc
Port 80 being the web interface and port 5555 being the "SCPI" command interface. Still lot of (undocumented?) ports open...
-
Yes, the same problem in mine :-//
Regards
-
But, this points to pretty much non-existent QA process at Rigol if this kind of issues get through to released firmware.
There are several 30+ page threads on this site that can attest to their level of QA. Granted they do seem to listen and at least try to fix some of the issues, but they tend not to release change logs with the new firmwares, so you can't always be sure what they fixed and what they didn't. Also new undocumented features/bugs show up in them sometimes. You get what you pay for...
-
Another note, there seems to be lot of ports open on these things (and even ancient Portmap/RPCbind service):
Don't worry, it's very unlikely that somebody will write an exploit for Rigol scopes.
99.99% of the exploits are written for one and the same OS from redmond (which is not used in Rigol scopes).
Anyway, you should never connect T&M instruments to a normal lan. Only to a lan that can not be accessed by other computers and is not connected to the internet.
Or use it only with computers that are not using an os from redmond.