02-27-2015 10:24 PM - edited 03-03-2019 07:46 AM
Hi Team,
While copying IOS into the switch via TFTP I was getting below notification.
"Address or name of remote host []? 192.168.102.21
Source filename []? c2960-lanbasek9-mz.150-2.SE7.bin
Destination filename [c2960-lanbasek9-mz.150-2.SE7.bin]?
Accessing tftp://192.168.102.21/c2960-lanbasek9-mz.150-2.SE7.bin...
Loading c2960-lanbasek9-mz.150-2.SE7.bin from 192.168.102.21 (via Vlan2): !!!!O!!!!!O!!!"
File is copying properly, checked MD5 Checksum also, but unable to understand this notification." O!!!!!O!!!"
After reloading the switch with new IOS, the switch is working fine as before.
Need to know, why we are getting this kind of notification?
regards,
Sacchidanand
Solved! Go to Solution.
02-27-2015 10:47 PM
The "O" means re-try. Happens when you've got a some congestion with your link or when the switch is having high CPU.
You'll see "O" when there's not enough flash space and the switch CPU shoots up because it's trying to find any space to put the file in.
As long as you've checked and compared the MD5 hash is identical, you should be right.
02-27-2015 10:47 PM
The "O" means re-try. Happens when you've got a some congestion with your link or when the switch is having high CPU.
You'll see "O" when there's not enough flash space and the switch CPU shoots up because it's trying to find any space to put the file in.
As long as you've checked and compared the MD5 hash is identical, you should be right.
02-27-2015 11:03 PM
Thanks Leo,
02-28-2015 09:27 AM
Hi Leo,
I seem to have observed this phenomenon very often on all at least decently recent IOSes for 2960 and alike switches. It was always in a lab setting, a single switch with no other connections or work to do whatsoever, directly connected to a PC running a TFTP server software, and the TFTP download producing exactly these kinds of timeouts. There was no reason for high CPU (although I haven't verified that to be honest), no congestion on the single 100Mbps link between the PC and the switch. Yet, I have been able to see these timeouts reproducibly. A 2950 Catalyst in the same setting will download its own IOS from the same PC and TFTP server just fine.
My personal gut feeling is that there must be some regression in the TFTP client code in these recent IOSes. Because of this, I have moved over to using HTTP for IOS downloads.
Best regards,
Peter
02-28-2015 02:37 PM
That's really weird stuff observed, Peter.
About 20 hours ago (0100 GMT, 27 February 2015), I had to do an IOS upgrade (using the TAR file) on a 2960S and using the FastEthernet0 port. I saw no such thing. For obvious reasons, CPU was high but nothing abnormal.
At work, I use ZeroTouch to build switches and also never observe such thing.
I'll have to take a look when I get back to work.
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide