Run an MX250 in a small office/data center.
Recently had our new network printers [Ricoh IM C2000s and IM C3000] installed. Updated from older Ricoh models that didn't have this issue [new ones operate on an Android /Linux hybrid OS, I've been told. Old ones were just OpenBSD.]
New printers are on same VLAN as clients, but printing is so slow that it either fails or takes hours to print. Spooling is a nightmare. After numerous visits from print vendor, we determined it was our network after hooking my PC and one Ricoh up to a switch and testing with no timeouts or errors.
Vendor's guess was some QoS feature squashing the print traffic. I'm not sure where to begin here. Has anyone else encountered this?
I can't wait till printing is obsolete.
We are migrating a site with a couple of Ricoh MFPs from Cisco IOS switches to Meraki MS210s tomorrow so I'll let you know how that goes... The devices we have are MP C5504ex MFPs. Not exactly the same but either way it may offer some insight...
I just replied to your post on the Meraki subreddit, but will share here too-
Just a guess, but the symptoms sound like it could be a duplex mismatch between the printer NIC and switch port. Have you tried checking that?
Hey, thanks for replying! I know the printer was set on auto, then defined at 100 when we were trouble shooting. The switch port on the MX is set to full duplex 1 Gb, but I'll need to dig into the Catalyst's CLI to check those ports.
show interface status
It will show speed and duplex with an a- if set to auto, the arrowed one below has a forced speed and duplex:
In this day and age, you should only be needing to use auto/auto. Make sure nothing is locked.
Does it affect machines printing in the same VLAN as the printer? If so, it has nothing to do with the MX. The traffic does not pass through it.
If it is only affecting traffic that is routed between VLANs with the MX check the security centre event log and see if anything is being triggered. Perhaps trying turning off AMP and IPS for 10 minutes and see if the behaviour is changed in any way.
Next try a packet capture. Anything interesting like retransmits, error, or anything like that?
Once we had completed our migration of the Cisco IOS switch stack to MS210s, we tested the printing and for us it has been fine.
Just now getting to circle back to this, but guess what I found on the Catalyst's port for one of the printers:
So, I changed the Ricoh itself to half duplex 100 Mbps and tried to print again. While still taking too long for 20 pages, the whole job completed which is way better then when I started!
Next move I guess would be to change the Catalyst int settings to auto/auto or at least full duplex?
It looks to me like the printer is hard coded to 100/full (which is why the Catalyst is reporting 100/half). If so you need to match the configuration.
I'd personally get the printers reconfigured to use auto/auto.
Would I need to change the interface settings on the Catalyst at all? Pretty sure the printer started out on Auto... changing it back now to see how it does with Auto Gigabit enabled
The switch and the printer need to have exactly matching speed/duplex settings. You need to change whatever is required to make that the case.
So, here's what I did:
Checked the int status of the Catalyst, saw it was reporting a-half, a-100
Changed the Ricoh to match that, tested printing [slow but job completed]
Changed Ricoh to auto-select, enable 1Gbps, testing printing [fast!]
Checked int status again on Catalyst, saw message on the console that the int had changed its state to "down" twice [%LINEPROTO-5-UPDOWN and %LINK-3-UPDOWN], then changed its state to up on both again
Reconnected to console, checked int status to see port reporting a-full, a-1000 now
Printing like a freaking champ at the moment
Sorry to falsely accuse the Meraki!