Rampage beachball

[TF019]

Hello,

I've recently moved from Trueflow to a new shop which runs Rampage 10.4.
Quite and experience for me as I had worked in Trueflow for 15 years.
I built my network from the ground up and had a very fast response time with my Trueflow server.

Anyway, when I work with the RAM client I'm experiencing extremely long waits in response to the RAM server. Getting the beach ball spinning on my Mac, the other two operators that have worked here for years get the same thing.

I'm not sure if it's a network switch or the server itself, I noticed the RAM server has not been defragged for quite some time so I am planning on starting there after a backup is completed which also has never been doen. They have a gigabit network with a HP Procurve 2824 switch. I also notice the the RAM server has two NIC ports on the back but they are only using one.

Anyone have an idea where I should start with this?

 

[Stiv]

10.6 here, are you waiting for the client to refresh? What are the refresh times set to in the prefs of the client?

Is the delay from when you click open the monitor proof to when it actually opens the proof or in TrapEdit?

How many pages are in your page list? When was the last time you rebuilt it?


I guess we need to know when is the actual delay where the beachball is spinning?

 

[TF019]

Yes, we are waiting for the client to refresh, refresh display is set to every 1 min, looks like it can go down to 30 seconds. The delay occurs for all processes from normalizing to RIP, FPO and proofs. Seems like it's more of general network issue as we also see beach ball when saving large photoshop files to the Rampage Mac volume. Some times when we try to open the client we experience huge delays.

If all 3 operators are doing a large task on the Rampage server we all experience the beach ball. The Rampage server also holds all Mac Volume files as well as Rampage data.

Things are crawling even worse now as I write this...

 

[TF019]

The other guys stated that the the page list has not been rebuilt in quite some time. We had the same feature in Trueflow where we would delete the log files on the server and also the clients. We will start with that when we get ready to head out today and see what happens in the morning. Also the defrag of the server is needed as well, this is planned for Friday end of day when we head out.

 

[Gramann]

If all else fails, try this product:

Mac Windows File Sharing - Connect Mac to Windows Server - GroupLogic

I have seen it work wonders.

 

[Stiv]

If you have 3 clients RIPping files at once - how many RIPs do you have? We have two, each with a prescan, RIP and one has the proof output. If we have 3 clients on the RIP at once everything crawls. IMO, you are not getting past this.
I usually have around 8,000-10,000 pages in the page list. I rebuild it at least a few times a week. Just hit the button when the last guy leaves for the day and start fresh.
Are you rotating the pages for use with Preps? If you are not then the RIP is working too hard on the impos. Need more info on this let me know.
You do realize that Rampage v10.4 is quite long in the tooth? The new v12 has each RIP on a quad core. All colors RIP simultaneously and the monitor proof is made at the same time. Crazy fast.

 

[TF019]

We rebuilt the page list last night and only came up with one or two errors. I don't think that was the issue. I noticed this morning when a GB file was being written to the Mac volume of the Rampage server that the client came to a crawl again. It seems to be more of a networking traffic issue than anything. It was interesting to read that when you have 3 operators Ripping at once everything comes to a crawl as that is what we experience.

I'm going back to what I said about the NIC cards, the server has two of them. If we could use one for Mac volume and one for RAM volume we would have two ports to read and write from.
Not sure if that is possible or not. Any thoughts?

 

[kansasquaker]

ExtremeZ-IP did wonders for the performance of our rampage v10 fileserver. If you don't have it, get it immediately!

I'm really interested in your idea of using both NICs. Our box is the same, and like you we're only using one of the NICs. But in our configuration JobVol and RamVol are both folders on the same logical drive. I would have no idea how to partition traffic based that in Windows. Like you, we have three ops, but don't have nearly the performance issues you're having. However, it would be nice to copy large files faster . . .

 

[TF019]

We are now considering setting up an older G5 and connect it directly to the Epson proofer with a USB cable vs. the ethernet cable, using a hot folder made from automator on the G5 we could just drag and drop a pdf from the proof folder to the hot folder on the G5, that way network traffic would be reduced so Rampage would not be tied up sending the proof, just making the pdf to drop into hot folder. Drag and drop.

I will check into ExtremeZ-IP, thanks

 

[jay12345]

We are now upgraded to v12 from v11 and have never experienced this issue. The speed increase with v12 is absolutely amazing, but even before the upgrade things ran rather smooth without any real holdups. v11 was running on a Quad Core Rip/Server combo and now that box is a dedicated server with the v12 rip running on the new box with dual 6 cores. I have never rebuilt the page list either. In fact i dont even know how to do it. Is this something you are supposed to do on a regular basis and if so how do you do it? Why do you do it? And why was i never told about it.

 

[Stiv]

Ours are all on dedicated boxes. I can't imagine that Rampage set-up your RIP, etc. all on the same server. Yikes, no wonder you are having issues. I haven't dealt with server issues in quite some time and will defer that advice to others.

Make sure that no pages or jobs are running, turn off any process, RPD, RIP etc. and under Utilities is Rebuild Pagelist (or something like that). If you have any un-deleted files from deleted jobs - this will allow you to remove them from the list. Will you be faster? I doubt it but your list will be spotless. I add around 10k pages per month and am constantly archiving and removing old jobs. I want my page list to be perfect.

 

[TF019]

Our Rampage RIP is a XEON 3.2Ghz with 1.5GB of RAM.
The RIP handles all RIP, FPO and PROOFS.
The file server which is separate hosts the Pathway archives and all mac volumes.
This means the file server is getting hit at the same time with large mac files as well as all RIP traffic.

Other than upgrading to a Quad or Six core box I don't see any options to increase network traffic. This seems like a flawed workflow design as the Pathway forces all traffic to one location where RIP and mac files are stored.

Anyone know of another option than upgrading our Rampage box to hold everything let me know.

 

[kart97]

One of the things that tended to slow the system down was having a ton of Font or image lists... to check if this is the problem... from the client go to the edit menu and check for either of these sets of lists... if there are any there from old jobs that are no longer in the system... delete them and the system will usually speed up.

 

[EZPower]

Mac server and never look back. We also managed to get multiple rampage boxes with one being a dedicated shooter, and two others being rip and trap stations. Fast. really fast.

 

[[email protected]]

I've been reading through this for a couple of days, and you really need to upgrade your rip box, and your Rampage version. ExtremeZ-IP is worth every penny as well. You are operating with software and hardware that is probably five or six years old. How old is that server?

We are running the latest version 12 on a fire core quad box, with another box for ffps, and a separate shooter. We seldom rebuild our page list, and have over 13,000 pages in it. We have 5 simultaneous operators. Rampage cost of ownership is low, but you still need to upgrade it like everything else. We try to upgrade hardware and software at least every three years. Upgrades are necessary to support new software versions, and maximize productivity.

 

[Bob888]

I had a similar problem to the one you described. Our system started to slow down after a couple of years of use. We are running Version 11 and our Rampage Volume resides on an Apple XServe. Talked to Rampage and they were not able to diagnose the problem. Digging around on the Rampage volume I happened to notice that Pathways was not removing the profiles when jobs were archived. Once I copied and then deleted all the unneeded Profiles off of the server, the server returned to its original speed. When archiving and restoring jobs moving the profile folder is done manually. If I had to guess, I don't think the XServe can handle thousands of tiny files.

 

[Chasfinch]

I think for speed you need to get the FIRECORE Rampage box. It made a huge difference for us.

 

[Raymond Ramirez]

The rampage volume has a net speed test that can be performed to see if its a Network or hardware issue. I Think it maybe hardware