[OSM-dev] Osmosis hourly replication broken?
brett at bretth.com
Mon Sep 13 00:43:37 BST 2010
Hmm, I can't remember details, but I probably did it during the large
re-factor for 0.36. I tried not to make too many code changes, but had to
change a few things in order to minimise dependencies once I separated code
into distinct projects. The sequenceNumber and timestamp fields are the
only ones that make sense for hourly replication anyway so it was always a
bit dodgy that I included them in the file.
Until the upgrade approx 12 hours ago, I was running 0.35 on the planet
server. Now I'm running the latest 0.37-SNAPSHOT which obviously
incorporates the 0.36 re-factor.
Apologies for the breakage, but I suspect the best solution is to upgrade.
On Mon, Sep 13, 2010 at 8:00 AM, Frederik Ramm <frederik at remote.org> wrote:
> currently, the file
> #Sun Sep 12 21:02:11 UTC 2010
> which makes my Osmosis 0.32 complain with
> java.lang.NumberFormatException: null
> at java.lang.Long.parseLong(Long.java:401)
> at java.lang.Long.parseLong(Long.java:478)
> where it expects a "txnMax" field to be present (and I'm using that version
> daily every day, and yesterday it still worked).
> The current minute-replicate file still looks right:
> #Sun Sep 12 21:53:05 UTC 2010
> Has there been any change to the hourly state.txt format specifically that
> would mandate an upgrade? Matt says it works for him with 0.36 but I can't
> understand why the minutely still looks like it used to be.
> Frederik Ramm ## eMail frederik at remote.org ## N49°00'09" E008°23'33"
> dev mailing list
> dev at openstreetmap.org
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the dev