Hi, i have a question about a problem of duplicate host.
One user report me this problem. He was attached only to seti@home, and before change the client (from cruncher to KWNS), he make a backup of the boinc folder. The update of the client failed (all wu in error because the app_info was not ok and boinc was unable to find the client). This is a screen of the error: http://img81.imageshack.us/img81/1217/errorib0.png
After this, he close BOINC and restore the backup. It works ok like before, but the scheduler generate a new host CPID :| and a new host in the account! Why this?? This is the log after restore of the backup:

27/03/2007 23.01.08||Starting BOINC client version 5.4.9 for windows_intelx86
27/03/2007 23.01.08||libcurl/7.15.3 OpenSSL/0.9.8a zlib/1.2.3
27/03/2007 23.01.08||Data directory: C:\BOINC
27/03/2007 23.01.08|SETI@home|Found app_info.xml; using anonymous platform
27/03/2007 23.01.08||Processor: 1 AuthenticAMD AMD Athlon(TM) XP 2000+
27/03/2007 23.01.08||Memory: 767.47 MB physical, 2.21 GB virtual
27/03/2007 23.01.08||Disk: 76.32 GB total, 51.14 GB free
27/03/2007 23.01.08|SETI@home|URL: http://setiathome.berkeley.edu/; Computer ID: 774216; location: home; project prefs: default
27/03/2007 23.01.08||General prefs: from SETI@home (last modified 2006-07-31 23:47:11)
27/03/2007 23.01.08||General prefs: no separate prefs for home; using your defaults
27/03/2007 23.01.08||Remote control allowed
27/03/2007 23.01.08||Listening on port 31416
27/03/2007 23.01.08|SETI@home|Resuming task 03no03aa.12667.9393.254826.3.8_2 using setiathome_enhanced version 519
27/03/2007 23.01.08|SETI@home|Started upload of file 27se04ab.11205.19489.248590.3.33_3_0
27/03/2007 23.01.08|SETI@home|Fetching scheduler list
27/03/2007 23.01.13|SETI@home|Scheduler list download succeeded
27/03/2007 23.01.18|SETI@home|Sending scheduler request to http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi
27/03/2007 23.01.18|SETI@home|Reason: Requested by user
27/03/2007 23.01.18|SETI@home|(not requesting new work or reporting completed tasks)
27/03/2007 23.01.23|SETI@home|Scheduler request succeeded
27/03/2007 23.01.23|SETI@home|Generated new host CPID: 0e75ed54f7c7d5eb4db7b45e19c0946a
27/03/2007 23.01.29||Project communication failed: attempting access to reference site
27/03/2007 23.01.30|SETI@home|Temporarily failed upload of 27se04ab.11205.19489.248590.3.33_3_0: http error
27/03/2007 23.01.30|SETI@home|Backing off 1 minutes and 23 seconds on upload of file 27se04ab.11205.19489.248590.3.33_3_0
27/03/2007 23.01.32||Access to reference site succeeded - project servers may be temporarily down.
27/03/2007 23.01.48|SETI@home|Started upload of file 27se04ab.11205.19489.248590.3.33_3_0
27/03/2007 23.02.09||Project communication failed: attempting access to reference site
27/03/2007 23.02.10|SETI@home|Temporarily failed upload of 27se04ab.11205.19489.248590.3.33_3_0: http error
27/03/2007 23.02.10|SETI@home|Backing off 3 minutes and 26 seconds on upload of file 27se04ab.11205.19489.248590.3.33_3_0
27/03/2007 23.02.11||Access to reference site succeeded - project servers may be temporarily down.
27/03/2007 23.02.58|SETI@home|Started upload of file 27se04ab.11205.19489.248590.3.33_3_0
27/03/2007 23.03.20||Project communication failed: attempting access to reference site
27/03/2007 23.03.20|SETI@home|Temporarily failed upload of 27se04ab.11205.19489.248590.3.33_3_0: http error
27/03/2007 23.03.20|SETI@home|Backing off 18 minutes and 3 seconds on upload of file 27se04ab.11205.19489.248590.3.33_3_0
27/03/2007 23.03.21||Access to reference site succeeded - project servers may be temporarily down.
27/03/2007 23.03.28|SETI@home|Sending scheduler request to http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi
27/03/2007 23.03.28|SETI@home|Reason: Requested by user
27/03/2007 23.03.28|SETI@home|(not requesting new work or reporting completed tasks)
27/03/2007 23.03.33|SETI@home|Scheduler request succeeded
27/03/2007 23.04.11|SETI@home|Started upload of file 27se04ab.11205.19489.248590.3.33_3_0
27/03/2007 23.04.33||Project communication failed: attempting access to reference site
27/03/2007 23.04.33|SETI@home|Temporarily failed upload of 27se04ab.11205.19489.248590.3.33_3_0: http error
27/03/2007 23.04.33|SETI@home|Backing off 5 minutes and 7 seconds on upload of file 27se04ab.11205.19489.248590.3.33_3_0
27/03/2007 23.04.34||Access to reference site succeeded - project servers may be temporarily down.
27/03/2007 23.04.52|SETI@home|Started upload of file 27se04ab.11205.19489.248590.3.33_3_0
27/03/2007 23.05.14||Project communication failed: attempting access to reference site
27/03/2007 23.05.14|SETI@home|Temporarily failed upload of 27se04ab.11205.19489.248590.3.33_3_0: http error
27/03/2007 23.05.14|SETI@home|Backing off 49 minutes and 28 seconds on upload of file 27se04ab.11205.19489.248590.3.33_3_0
27/03/2007 23.05.15||Access to reference site succeeded - project servers may be temporarily down.
27/03/2007 23.06.27|SETI@home|Started upload of file 27se04ab.11205.19489.248590.3.33_3_0
27/03/2007 23.06.48||Project communication failed: attempting access to reference site
27/03/2007 23.06.48|SETI@home|Temporarily failed upload of 27se04ab.11205.19489.248590.3.33_3_0: http error
27/03/2007 23.06.48|SETI@home|Backing off 2 hours, 25 minutes and 29 seconds on upload of file 27se04ab.11205.19489.248590.3.33_3_0
27/03/2007 23.06.50||Access to reference site succeeded - project servers may be temporarily down.
28/03/2007 0.28.58|SETI@home|Started upload of file 27se04ab.11205.19489.248590.3.33_3_0
28/03/2007 0.29.55|SETI@home|Finished upload of file 27se04ab.11205.19489.248590.3.33_3_0
28/03/2007 0.29.55|SETI@home|Throughput 2895 bytes/sec
28/03/2007 1.10.54|SETI@home|Sending scheduler request to http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi
28/03/2007 1.10.54|SETI@home|Reason: Requested by user
28/03/2007 1.10.54|SETI@home|Reporting 1 tasks
28/03/2007 1.10.59|SETI@home|Scheduler request succeeded
28/03/2007 1.10.59|SETI@home|Message from server: Can't find host record
28/03/2007 1.10.59|SETI@home|Generated new host CPID: bc07657a8b3d56cab08100c48b4b0f94

The original host (774216) exist in the account page and seems to be ok. Why boinc generate other host and CPID? I want to understand....:P

---

This does tend to happen after restoring backups, it's harmless.

I seem to recall someone posting that it was due to a sequence number somewhere - if the server receives a number it's already seen, it generates a new host ID. My recollection may be wrong, of course.

---

The original host (774216) exist in the account page and seems to be ok. Why boinc generate other host and CPID? I want to understand....:P

This is quite normal. How it works:
when client (eg. your home box) connects project server, it identifies itself using host identity. It also includes the number of connection attempts so far. These two numbers are stored in client_state.xml for each project separately - look for tags <hostid> and <rpc_seqno>. The later is really culprit of your problems: project server also stores this number and expects that upon subsequent successful connections, sequence number from client side should be larger than the number stored on server. If every connection is successful, it will be larger by exactly 1. If client made some connection attempts that failed (eg. during project server outage), then clients sequence number could be significantly larger than the number stored on server. If everything is fine, then clients sequence number is never lower than the number stored on server.

In your case it went wrong. The user made backup, then client made successful connection, things failed and user restored backups. This made clients sequence number lower than the number stored on server.

When things are not OK, project server creates new host ID as if your computer never had one before.

After such happening the only thing you can do is merge the two hosts on projects web pages. The effect of this is that you free up some space on projects database and the stats for your box will reflect also the past work. It will continue to use the new host ID though.

Some projects (notably Einstein@Home) will re-issue work units that get lost during mishappening to such merged host but mostly they won't.

---

Metod ...

Thanks very much to MikeMarsUK and Metod, S56RKO for the information about this case :)

Bye

---

"After such happening the only thing you can do is merge the two hosts on projects web pages."
I have a slightly different problem. I got one host-ID when I started BBC. Later I started another project via Boinc, and got a new host-id. I am no longer engaged in the BBC's project - how can I merge these 2 ID's ?

---

"After such happening the only thing you can do is merge the two hosts on projects web pages."
I have a slightly different problem. I got one host-ID when I started BBC. Later I started another project via Boinc, and got a new host-id. I am no longer engaged in the BBC's project - how can I merge these 2 ID's ?

You always get different host IDs if they are different projects (unless you're extremely lucky and get the same number :D), and you can't merge them because they are completely independent databases.

Thanks Nicolas for your input. I guess I used incorrect terminology - Boincstats shows me with 2 cpus:-
[ url ]http://www.boincstats.com/stats/boinc_host_stats.php?pr=bo&st=0&userid=24287861d04f51a94a4a1ff73a2a1305[ /url ]
- the first has BBC, the second has 2 other Boinc projects.
However, Boinc combined statistics has the same cpu-id for all three projects! So maybe it's just taking time for the information to get fully distributed - it was only a couple of days ago that I synchronised names etc.

---

Ahh that... Well I have a dozen of those, even though I have only two computers, and I don't know how to merge them :(

Hi Nicolas. I registered again with BBC via BOINC, and sure enough got my 'new' cpu registered alongside the old one. I was able to merge them internally in BBC (I think the old one gets merged into the new), but this info doesn't seem to get distributed/updated in e.g. boincstats. I guess it's something one just has to live with.

---