Stanley I don't have a lot of experience with Veritas Netbackup, but if it's like some of the other Veritas products I have been exposed to in years past, the problem is probably with it rather than with the network. My first guess would be that the Veritas end of it is configured incorrectly, and that there is some timeout setting that isn't set right. If you are using some Veritas agent to backup a live database, I'm about 90% sure that the problem is with that agent. We'll assume that isn't it, and we'll also assume that all the obvious causes (power surges, full backup media, open files, someone manually stopping the backup, a full swapfile, database errors, hardware faults, inadequate memory, crashed processes, etc.) have been ruled out as well. If they haven't been ruled out, do so before you blame the network; if you can get a partial backup, the network is probably OK. The next thing that I would do is take a look at the saturation on the network segment that your Linux server is on WHILE the backups are timing out. If there is too much traffic, odds are the packets are getting dropped by the network and they just aren't getting from the Linux box to the Veritas one fast enough because of the interference. The easiest way to check the saturation is to take a look at the hub/switch collision lights and see if they are going off like crazy, but using a software diagnostic (like Ethereal or Shomiti) would probably be a better idea. I would guess that there are probably several nodes on the network being backed up at the same time (all pushing a LOT of data), and that could be the cause of the latency. I've run into this problem myself on at least one occasion -- all I had to do to solve it was stagger the backup times. Getting on to your question, most 10/100 cards have an extra (or different) connect light that goes on when they get a full 100Mb connection. Most hubs and switches behave in the same manner. Check either one at any time and see how fast you are connecting. If you are in such a position that neither one is giving you any indication, you can always plug into the diagnostic port on the hub/switch and take a look from there. If you are using some of the fancy Cisco stuff out there, you can probably even take a look at the port characteristics through a browser. Here's hoping this helps! Kevin Finch Network Administrator DST Systems, Inc. 816/435-6039 krfinch@dstsystems.com stanley.g.martin@mail. sprint.com To: kclug@kclug.org Sent by: cc: owner-kclug@marauder.i Subject: Backup Problem lliana.net 01/04/2002 09:01 AM First Post I'm running RH7.1 on my server and I am experiencing a lot of trouble getting my backups to work. My server is in a corporate Datacenter and backups are controlled by a group that uses Veritas Netbackup. Ever since we installed this box, backups have never worked consistently. Our network group has set up their switches to run at 100MB Full Duplex. How can I make sure my NIC is running at this speed and if not, change it? The backup will run for a while and then time out. I'm not saying the problem is my network setting, but neither Veritas or the backup group can come up with anything better. Stanley G. Martin Midwest Consulting Group Sprint Platform & Strategy Mgmt 913.315.3133 Stanley.G.Martin@mail.sprint.com This e-mail and any attachments are intended only for the individual or company to which it is addressed and may contain information which is privileged, confidential and prohibited from disclosure or unauthorized use under applicable law. If you are not the intended recipient of this e-mail, you are hereby notified that any use, dissemination, or copying of this e-mail or the information contained in this e-mail is strictly prohibited by the sender. If you have received this transmission in error, please return the material received to the sender and delete all copies from your system.