|
osdir.com mailing list archive |
|
Subject: [OSM-dev] GPX Import Daemon - msg#00684List: gis.openstreetmap.develbeen processed for about a week now. SteveC has pushed out several rounds of updates to it from me over the last few days, and has made sure the daemon is running, but there is still no sign of it actually processing any logs. Needless to say it appears to work just fine in my test environment... I think somebody with access to the server needs to have a look at production.log for lines starting with "GPX Import" to try and see what it is doing - what we should be seeing is an "importing foo.gpx" message as it processes each log. If instead we are seeing a "daemon wake" message every fifteen minutes then it obviously thinks it has nothing to do for some reason. Using strace to see if the server is wedged, and if so what it is doing, might also be helpful to try and work out what is happening. Tom -- Tom Hughes (tom@xxxxxxxxxx) http://www.compton.nu/ Thread at a glance:
Previous Message by Date: click to view message previewRe: [OSM-dev] Availability of slippy map during LinuxTag 2007On Wed, 2007-05-30 at 18:42 +0100, Chris Jones wrote: > On 30 May 2007, at 12:18, Jon Burgess wrote: > > > I've been collecting some simple stats from the access blocks over the > > past week. We typically serve around 1M tiles per day to a couple of > > thousand different IPs with only 1 - 10 people hitting the 10k limit > > on any day. > > Does this take into account ISPs that force all http traffic though > one or two 'transparent' proxies? > > Virgin/NTL/Telewest for example do this - http:// > homepage.ntlworld.com/robin.d.h.walker/cmtips/trancache.html provides > a comprehensive list. > -- No, there is nothing done to take account of these transparent proxies. There is a simple count of tiles served per IP address. If another user of the proxy exhausts the tile limit then you are SOL. If and when we get a white listing system in place this could help. Thanks for the pointer to the info. Jon Next Message by Date: click to view message previewRe: [OSM-dev] Disappearing RoadsHmm... It looks like this is happening again, only in a different way: now a few streets I edited in potlatch last night show up fine in both potlatch and josm, but have sections missing on the newly-rendered Mapnik map. Have a look at way 4265058 located in the approximate bounding box defined by: 41.7748, -87.9966 41.7579, -87.9879 The area in question is also visible at: http://www.informationfreeway.org/?lat=41.766326094672515&lon=-87.99386025930893&zoom=15&layers=0000F0B In the above view there are also a few streets to the west of that section that are missing pieces. On 5/21/07, Cory Lueninghoener <cluening@xxxxxxxxx> wrote: > Last night I was playing with potlatch on downtown Chicago (somewhere > around 41.8822, 87.6295), and when I was done I wanted to clean up a > few things in JOSM. When I downloaded the area, I found that many of > the already-existing ways and segments (perhaps points too) in the > area had disappeared. I tried the same with the josm-latest from last > night and saw the same thing. Oddly, all of the parts still show up > in potlatch, and they still do this morning on a different computer > while JOSM still can't see them. Is this another known problem > related to the 0.4 API upgrade, or is something more sinister going on > here? > > I'm relatively new to the OSM world, so I'm not quite smart enough to > track the problem down further at the moment, but I'll happily lend a > hand as needed to keep my hard work from disappearing. > > -- > Cory Lueninghoener > Perl, C, & Linux Hacker > http://www.wirelesscouch.net/~cluening/ > -- Cory Lueninghoener Perl, C, & Linux Hacker http://www.wirelesscouch.net/~cluening/ Previous Message by Thread: click to view message preview[OSM-dev] What do we do with tags with empty keys or values?Hi, Todays planet.osm contains 1700 objects with empty tag names (446 with tag k="" and 1239 with tag k=" "). There are far more empty values, though, grepping for v="" gives me 67109 hits. What should we do with those tags? Is there any use for them or could they be safely removed? If so, would it be better to remove them via the API or directly via SQL access to the database? Regards, Hakan -- The key to immortality is first living a life worth remembering... Next Message by Thread: click to view message previewRe: [OSM-dev] [OSM-talk] Oh dear - planet has duplicate id'swow - and you know what, it's true - somehow there is no primary / uniqueness constraint on that id: http://trac.openstreetmap.org/browser/sites/rails_port/db/create_database.sql#L69 same goes for current_nodes - compare with current_ways (line 110) which has appropriate primary key declaration. i see you've filed a ticket on this. perhaps we should move any further discussion to dev@xxxxxxxxxxxxxxxxxx cheers, dan. ----- Original Message ---- From: David Earl <david@xxxxxxxxxxxxxxxxxxxx> To: OSM <talk@xxxxxxxxxxxxxxxxx> Sent: Wednesday, May 30, 2007 6:42:54 PM Subject: [OSM-talk] Oh dear - planet has duplicate id's Oh dear, today's planet file contains two pairs of duplicate segments: they have the same id. Surely id's are unique, yes? Isn't that the point? It doesn't look like they are new, but it wasn't like this last week or I'd have noticed. (There may be others, but my program crashed trying to do an insert on a duplicate primary key) Excerpt below... 18591604 and 18591605 are the culprits, and this is a contiguous section of the file. (Also, note that 18591606 doesn't have any useful tags and 18591607 is empty.) David ---------------------- <segment id="18591603" from="22542877" to="22542874" timestamp="2007-05-11T08:24:21+01:00"> <tag k="created_by" v="almien_coastlines" /> <tag k="source" v="PGS(could be inacurately)" /> </segment> <segment id="18591604" from="22542876" to="22542878" timestamp="2006-12-31T00:20:10+00:00"> <tag k="natural" v="coastline" /> <tag k="created_by" v="almien_coastlines" /> <tag k="source" v="PGS" /> </segment> <segment id="18591605" from="22542869" to="22542875" timestamp="2006-12-31T00:20:10+00:00"> <tag k="natural" v="coastline" /> <tag k="created_by" v="almien_coastlines" /> <tag k="source" v="PGS" /> </segment> <segment id="18591604" from="22542876" to="22542878" timestamp="2006-12-31T00:20:10+00:00"> <tag k="natural" v="coastline" /> <tag k="created_by" v="almien_coastlines" /> <tag k="source" v="PGS" /> </segment> <segment id="18591605" from="22542869" to="22542875" timestamp="2006-12-31T00:20:10+00:00"> <tag k="natural" v="coastline" /> <tag k="created_by" v="almien_coastlines" /> <tag k="source" v="PGS" /> </segment> <segment id="18591606" from="22542879" to="22542877" timestamp="2007-05-11T08:24:37+01:00"> <tag k="created_by" v="almien_coastlines" /> <tag k="source" v="PGS(could be inacurately)" /> </segment> <segment id="18591607" from="22542173" to="22542182" timestamp="2006-12-31T00:20:10+00:00"/> <segment id="18591608" from="22542880" to="22542881" timestamp="2006-12-31T00:20:10+00:00"> <tag k="natural" v="coastline" /> <tag k="created_by" v="almien_coastlines" /> <tag k="source" v="PGS" /> </segment> _______________________________________________ talk mailing list talk@xxxxxxxxxxxxxxxxx http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk ____________________________________________________________________________________Looking for a deal? Find great prices on flights and hotels with Yahoo! FareChase. http://farechase.yahoo.com/ Loading Comments...
|
|
