Suppose I start out with a single, standalone Zope server, and later realize I want to utilize the scaling capacity of ZEO. Here's how I would naively assume this would work: 1. Setup ZEO on another server (I know theoretically I can have ZEO on the same server). 2. Start ZEO. 3. Stop Zope on ths standalone server. 4. Backup Data.fs, place a copy on the ZEO server. 5. Place a custom_zodb.py in the INSTANCE_HOME folder of the standalone Zope server; the custom_zodb.py points to the ZEO server. 6. Start Zope on the previously standalone server. Is that all there is to it? I'd be able to answer this question myself if I could get ZEO running. In the meantime, I'd very much appreciate any insights into this seemingly common use case. Thanks, // mark -
Mark McEahern wrote:
1. Setup ZEO on another server (I know theoretically I can have ZEO on the same server). 2. Start ZEO. 3. Stop Zope on ths standalone server. 4. Backup Data.fs, place a copy on the ZEO server. 5. Place a custom_zodb.py in the INSTANCE_HOME folder of the standalone Zope server; the custom_zodb.py points to the ZEO server. 6. Start Zope on the previously standalone server.
Is that all there is to it?
Pretty much, but don't start ZEO until after step 4. Also, make sure your ZEO connection is secure -- if both machines aren't on the same well-firewalled network use a VPN, stunnel, or similar. Always always ALWAYS make sure that ZEO is stopped before trying to restore/replace the Data.fs. This isn't special to ZEO (you should do the same with a standalone Zope) but it's very easy to shut down your Zope and forget that the ZEO server is still running. Cheers, Evan @ 4-am
participants (2)
-
Evan Simpson -
Mark McEahern