SLC 5/05 Online Edit Resource cannot be obtained

This heavily loaded SLC kicked me offline during an online edit and now refuses to cancel or accept the pending edits. This is not the first time this has happened with this particular SLC 5/05. In the past, we have cycled power and downloaded the program to resolve this.

The processor status (structured) shows S:33/11,12 "Online edits disabled"

Changing the radix to binary:
S:33/11=1, S:33/12=0

I've read some tech notes on this issue, but I am still left with a question:

What would happen if I manually clear the status bit S:33/11 while the program is running?

I really don't want to shut down the whole area to follow the recommendation to cycle power to the SLC or download the program.

Thanks,
Paul

Nevermind. I got brave and tried toggling those bits, but they did not change...

I guess I will come in early tomorrow and cycle power before production starts...



Paul

I think (I don't have the software in front of me) that is just reading the status of edits in the controller. Changing the bit would not actually affect anything. The bit might change, but it doesn't affect the status of edits in the controller and the bit would toggle back off.

OG

** edit ** Too slow typing apparently.

I "cure" this one by deleting/reinstalling the ethernet driver in rslinx and it goes online fine after. This one is almost routine to me these days...

Same thing happens to me quite frequently.
Seems to happen on the 5/05 series, Ethernet communications, doing online edits.
It seems to lock it up to the point that there is no return, and the only method to get rid of the edits is to delete them.
If you apply the edits once again, it fails and boots the PC offline.
You have to download the application in order to get back to where you can complete online edits again.
If someone can find out what is really causing this problem, it would be much appreciated.
OkiePC and myself cannot be the only ones having this problem....

SNK said:

OkiePC and myself cannot be the only ones having this problem....

Click to expand...

I also have the same problem with just one SLC 5/05. None of the other six SLC 5/05s on the network have the problem and none of the PLC/5s, CLXs, or MLXs have the problem. Our other engineer also has the same problem with the same SLC.

Per knowledgebase technote 32963:
In RSLinx, open "Configure CIP Options" from the Communications menu. Under "Messaging Connection Parameters" set the dropdown box to "8.00 sec. retry interval, 256.00 sec. timeout". Set the "Max. Fast Messaging Connections per PLC" to 1 and the "Max. Slow Messaging Connections per PLC" to 1. The "Max. Unconnected Messages per PLC" can remain at 4.

Since I followed these instructions, I have not had the problem described above.

Thanks. I knew about that for the ML1100, but did not know that it also applied to the SLC 5/05. The thing is, I've got six other SLCs that don't exhibit any issues. I'll give it a try next time around.

How close to full memory is the processor?
I had a 5/04 that was pretty full and when i tried to do an online edit larger than 1 rung, it would kick me offline. Small edits were OK. I wonder if the same is happening here?

I have had the same problem. It comes and goes with various 505's on my network. The thing i notice is that when it happens with a processor i can go to rs linx and instead of the processor showing up as a rack icon with the project name under it , it shows up as just the processor icon and instead of the project name it shows the slc processor model #. If i open a 505 in rs500 that is having a problem and go to controller properties and click on the controller communications tab the route shows cip path and a bunch of #'s seperated by periods, on processors that isn't having problems the same tab shows the route as local and the IP address. On my plc network now i have 10 505's and 4 of them are showing up the way that causes online editing problems. The funny thing is that they seem to fix themselves over time.

We had already set up the CIP options according to the recommendations by A/B. It helps, but does not cure the problem we have.

This particular SLC has done this to us several times over the past two years. Each and every time it happens, about 8 hours later, the PLC will lose its program and require downloading. This is repeatable, and something that I had forgotten about since my counterpart here normally takes care of this system. This is the first time it happened to me.

Yesterday, he had the day off so I made the necessary changes online. I edited about 8 rungs with no problems to get a new blender running correctly, but needed to change the direction of a pair of agitator paddles. That change only required four rungs to be edited, but after editing two of the rungs and accepting them I was booted offline. Their blender still functioned okay, so it was something I could postpone.

So, I gave up trying to fix it without shutting down the whole operation and was resigned to doing it this morning. But, last night, about 8 hours after it first lost comms during the edit, they had to call my counterpart at home when their touchscreens stopped working.

He went into RSLinx and right clicked to view data tables (the quickest way to see if comms are working.) He said he only got three files, and when he tried to go online he got a message requesting a download or restore from EEprom. He downloaded the latest file and got them going.

I don't know if anyone else experiences the SLC dies 8 hours later part of this ordeal, but there is definitely something wrong with the SLC firmware IMHO.

THe details: 1747-L553C 5/05 CPU-64K Mem. OS501 Series C FRN 10
Revision (S:62)=5
Ethernet Daughterboard Series (S:9)=3
Ethernet Daughterboard FRN (S:10)=46

173 Program files
254 Data files
29493 instruction words used - 30658 data table
Memory left: 3259 instruction words

Paul

Paul, I've never seen the issue on ours with it dropping memory 8 hours later.

The one that gives us trouble is a 5/05 32K of memory and its got plenty of memory left, in fact we aren't even using 1/3 of it. It also appears that you have plenty of memory left.

I checked and my CIP settings are also the same as recommended in the KB.

I've been playing with it a bit and I've stumbled on something interesting. When I have the system comms set up to use an Ehternet/IP driver (AB_ETHIP-1) I get booted off when I do an online edit. However, if I switch and use the plain jane ethernet driver (AB_ETH-1) where I enter the PLC IP address into the table, I apparently can successfully make online edits. I just tried one and didn't get booted. When I switched back to the Ethernet/IP driver to undo the online edit, I got booted out.

edit to add: Well, maybe not. I was demonstrating this to my cow-orker just now and it didn't work. I got booted with both drivers.

Yeah, we have seen it with both driver types. Also, I should point out that it has happened with two different processors in the same application.

My counterpart says that if he waits about 30 seconds after each online edit transaction (ie. accept a rung, wait, accept another rung, wait, test, wait, assemble...) then it does not happen.

I am still dumbfounded by the 8 hours later it quits part, and curoius if any others have had that happen.

I posted a link to this thread at the ab forum but it has only had 11 views so far. Hopefully, R/A will come up with a fix in a future firmware release.

Paul

Paul, is the problem SLC5/05 doing any messaging?

Ours is sharing info with two PLC/5s through an ENBT/DHRIO bridge in a CLX rack. If my memory is correct (which is becoming more dubious) we never ran into this issue before we started sharing production data among the machines. We aren't messaging huge amounts of data, just a half dozen integers.

None of the non-problem SLCs are doing any messaging.

Also it would be nice to hear from KentuckyTech, SNK, and Blijj if their apps are also doing any messaging.

No MSG instructions. It does have two devicenet scanners handling a bunch of powerflex drives. It is also being polled by three HMI (FTView SE Standalone) as well as RSSQL.

The ethernet status usually shows 15 or 16 connections. There are zero errors/collisions/storms in the status monitor.

The port is fixed at 100Mbs Full Duplex and the switch port it's connected to is also fixed at the same settings.

avg scan time is ~70ms,

EDIT: There is a Controllogix PLC polling (MSG reads) it and another 5/05 reading a few integers too.


Paul