I manually updated Predictor@home from BOINC Manager (5.10.28 & 5.10.30) several days ago. After doing that, I found out that the project is temporarily down because it's moving. BOINC is still trying to communicate with this project. How do I get it to stop?

---

The only way is to detach from it, else BOINC will try to communicate with the project's scheduler every 24 hours (and defer communications for the rest of that time).

I've got the same bit going with Orbit@Home. No New Tasks and Suspend doesn't stop BOINC from trying to reach the scheduler, until it's gotten contact at least once.

---

Could this be done by editing XML? i.e., modifying the deferral time to something larger.

---

Oooh. I don't know. Perhaps that it works when you increase the number in between the <request_delay></request_delay> tags in sched_reply_*.xml.

It would be best to stop BOINC before it and restart it afterwards, then see what it does.

---

Oooh. I don't know. Perhaps that it works when you increase the number in between the <request_delay></request_delay> tags in sched_reply_*.xml.

sched_reply isn't ever read again after the request is parsed by the client.

---

Ah OK, so that's a no, then. :-)

---

How about min_rpc_time in client_state.xml? This is present in the project section. (Note that mucking about in this file could corrupt your boinc installation).

<project>
    <master_url>http://www.apsathome.org/</master_url>
    <project_name>APS@Home</project_name>
    <symstore></symstore>
    <user_name>MikeMarsUK</user_name>
    <team_name>BOINC Synergy</team_name>
    <host_venue>home</host_venue>
    <email_hash>xxxxxxxxxxxxxxxxxxxxxxxx</email_hash>
    <cross_project_id>xxxxxxxxxxxxxxxxxxxxxxxxxxxx</cross_project_id>
    <user_total_credit>6238.234740</user_total_credit>
    <user_expavg_credit>0.067288</user_expavg_credit>
    <user_create_time>1181325324.000000</user_create_time>
    <rpc_seqno>6787</rpc_seqno>
    <hostid>579</hostid>
    <host_total_credit>2792.693605</host_total_credit>
    <host_expavg_credit>0.080248</host_expavg_credit>
    <host_create_time>1181326328.000000</host_create_time>
    <nrpc_failures>10</nrpc_failures>
    <master_fetch_failures>0</master_fetch_failures>
    <min_rpc_time>1199963392.414257</min_rpc_time>
    <next_rpc_time>0.000000</next_rpc_time>
    <short_term_debt>0.000000</short_term_debt>
    <long_term_debt>55155.034253</long_term_debt>
    <resource_share>120.000000</resource_share>
    <duration_correction_factor>0.810181</duration_correction_factor>
    <sched_rpc_pending>0</sched_rpc_pending>
    <send_time_stats_log>0</send_time_stats_log>
    <send_job_log>0</send_job_log>
    <ams_resource_share>0.000000</ams_resource_share>
    <scheduler_url>http://www.apsathome.org/APS_cgi/cgi</scheduler_url>
    <code_sign_key>
xxxxxxx
.
</code_sign_key>
</project>

---

What exactly does the min_rpc_time represent? Is it a length of time (like 24 hours) or a specific time (like 9:30 am)?

---

What exactly does the min_rpc_time represent? Is it a length of time (like 24 hours) or a specific time (like 9:30 am)?

It's a specific time, in seconds since January 1, 1970 0:00:00.

---

I increased the min_rpc_time to several weeks. The countdown in the Projects tab increased accordingly. Let's hope it works.

Thanks for the help, guys.

---

The only way is to detach from it, else BOINC will try to communicate with the project's scheduler every 24 hours (and defer communications for the rest of that time).

I've got the same thing with LHC too.

If you suspend & BOINC still tries to communicate every 24 hours ... who cares??
Unless you are paying for every packet.

The repeatdly every few minute failed communication did drive me crazy.

---