Has anyone seen where you using software distribution to copy a sup720 advanced services IOS to a 6500 and it fails after copying 25 megs. The IOS is about 75 megs and the protocol used is TFTP. We were running RME 4.0.5 on LMS 2.6 but just upgraded to 4.0.6.
I have not seen a transfer stop after 25 MB. Most TFTP servers have a built-in limit of 32 MB, but our servers do not. However, you must be running recent device code as well to workaround the limitation. What version of code is currently running on the 6500? What platform is LMS running on?
LMS 2.6, RME was on 4.0.5 but I have just upgraded it to 4.0.6. Yes, you are correct, it would die at 36 meg. The IOS we trried to copy to the 6500's was 12.2(18)SXF7. That IOS is huge!!!
The image that is running on the 6509 is s72033-advipservicesk9_wan-mz.122-18.SXF7.bin. The platform is LMS 2.6. Also at the time we were on Common Services 3.0.5 and RME 4.0.5. Now we are on RME 4.0.6 and CS 3.0.6. Do I have to edit a file somewhere within Ciscoworks to fix the problem with TFTP timing out at 32 megs?
Okay. We started shipping out tftp server for Solaris in CS 3.0.6. However, you need to make sure you're using it. Look at /etc/inetd.conf. Make sure the tftpd line looks like:
tftp dgram udp wait root /opt/CSCOpx/bin/in.tftpd in.tftpd -s /tftpboot
If not, change it, then restart inetd.
The line was remed out. So I deleted the pound sign from in front of it. Now my name was:
tftp dgram udp6 wait root /usr/sbin/in.tftpd in.tftpd -s /tftpboot
I chaged it to:
tftp dgram udp wait root /opt/CSCOpx/bin/in.tftpd -s in.tftpd -s /tftpboot
Now how do you restart inetd?