On 04/03/14 20:39, R S wrote: > A customer of mine runs MX960 (dual RE), SRX5800 and SRX3600 (cluster with > dual RE), EX8200 (dual RE) and EX4200 (in VC), in the weekend will run a > disaster recovery test powering off those devices. > I’d like to avoid any problem with files corruptions, which is the safer > command to use to save current file systems of those devices family ? > > request system snapshot <enter>
Snapshots won't help this. Snapshots aid recovery, they don't prevent corruption. > Can I ask to run this command in production time on all the mentioned devices > ? Sure, on the EXen it can be a little slow, but should be 100% safe. Remember to run it on both RE's on dual-RE platforms. You may need "slice alternate" on the end of the command for the EXen, and the exact command depends on JunOS release. > Any additional suggestion ? Make sure you actually have matching JunOS on all RE's in a chassis/cluster/VC. Ditto any scripts commit, event, etc. If you're using fxp0 ensure both RE's have it patched. Ensure "system commit synchronise" is set on all dual-re platforms. Make sure rancid (or your equivalent) is actually working and has current configs for all devices, sadly there's still a good chance of corruption needing a rebuild on EX[34]200 after power loss.
signature.asc
Description: OpenPGP digital signature
_______________________________________________ juniper-nsp mailing list [email protected] https://puck.nether.net/mailman/listinfo/juniper-nsp
