changing to the new router on Friday night

| | Comments (5)
In the process of moving the xen dom0s from SVTIX to Market Post Tower, we have been using sphinx.prgmr.com for routing at both data centers. We have a  router setup now at Market Post Tower also (shinjuku.prgmr.com) and on Friday night we plan to move the gateway addresses for the xen subnets to the new router.

The downtime for this should be separate for each subnet, and should only last a few minutes. Once that is done, we will move the address that our upstream routes to over to shinjuku and route the remaining subnets for colocation back to sphinx at SVTIX. This will mean downtime for everyone at SVTIX and Market Post Tower, and should also only take a few minutes. We will then reboot shinjuku to make sure all the settings are saved correctly.

After this is all working, we plan to start using BGP at Market Post Tower, and eventually also SVTIX. We're planning on having additional transit providers at both data centers also, so it should help with the problems caused by the XO leased line earlier this year. Thanks, and please email support@prgmr.com if you have any questions.

we plan on starting this at 21:00, and only the network will be disturbed;  all guests will remain unmolested (save for a short network outage)


This went badly.   Shinjuku didn't start again after it had the new configuration, so we rolled back to the old router.  sorry.

5 Comments

Does prgmr publish information regarding its network that would allow measuring the number of hops and latency to its VPS services? For some applications, this type of measurement might help in choosing one VPS provider over another.

Browsing through old blog entries, I think I could get a good impression of the prgmr network, but having a single page where prgmr maintains this information current would be convenient.

you really want to do that yourself; conflicts of interest aside, the hops and latency to the vps depends entirely on hops and latency to where. If you measure us from a linode in the he.net fremont 1 data center, or anything else in the silicon valley area, we are going to look damn good. If you measure us from England? well, not so much.

I think I need to clarify. I was not trying to suggest prgmr should publish latency and number of hop values. Rather, I was thinking it might be helpful to publish information on its network, like the IP addresses of its gateways. Looking at recent blog posts, I see different networks assigned to prgmr, for example:

sphinx.prgmr.com 72.13.95.0/24
bull.prgmr.com 64.71.167.0/24
crock.prgmr.com 65.49.73.0/24

It looks like all these are all located at the same place. If all the VPSs are at this location it's easy to determine the latency from anywhere on the Internet to prgmr, but I'm not clear on whether VPSs are necessarily hosted at that location. If you still think there's a conflict of interest, perhaps you could spell it out, as I don't see it.

Examples of VPS providers publishing network information:
http://bitfolk.com/customer_information.html#toc_1_Network_settings
http://www.bytemark.co.uk/company/network

oh, right. that's not a conflict of interest, of course. I was thinking you were asking me to setup some sort of monitoring thing for people to see the hops to prgmr.com vs. linode or slicehost or whatever from various points;

You can all my IP addresses here:

http://whois.arin.net/rest/org/PRGMR/nets

except for 38.99.2.0/24 isn't there (that's in Sacramento, and we're not adding new people there.)

the networks prefixed with LIS are in he.net fremont 1, and the rest right now is in market post tower in san jose.

Thanks, that's helpful.

Leave a comment

About this Entry

This page contains a single entry by nick published on September 21, 2011 11:13 PM.

crock soft lockup. was the previous entry in this blog.

soft lockup on crock is the next entry in this blog.

Find recent content on the main index or look in the archives to find all content.