Hi all,
We've spun up a new RHEL7 Spectrum 10.4 SS server and all works fine except NCM capture for some devices. Never had this issue on our current Windows install but that's using the s'winds TFTP server :)
So, we're running the tftp-server (5.2-22) under xinetd (2.3.15-13) with all the right settings and most of our Cisco devices capture configs fine.
The ones that fail have a locally set tftp blocksize of 8192 for legacy reasons. I tried to set the in.tftd's blocksize in the /etc/xinetd.d/tftp file way up to 12k as below [way under the 65464 maximum but enough to satisfy these devices]
server_args = -c -s /tftproot -u svc-spectrum --blocksize 12288
But while the xinetd/tftp restarts and still works for the devices using the default blocksize anything set with a blocksize larger than 1468 fails. I've tried larger (32768) and smaller overrides (2048) while changing the devices to match, as well as changing the server_args from --blocksize to -B but no luck.
What do you do if running under RHEL to make this work? I can't be the only one.
Regards, JB.