boinc/sched/wu_check.cpp

168 lines
4.8 KiB
C++
Raw Normal View History

// This file is part of BOINC.
// http://boinc.berkeley.edu
// Copyright (C) 2008 University of California
//
// BOINC is free software; you can redistribute it and/or modify it
// under the terms of the GNU Lesser General Public License
// as published by the Free Software Foundation,
// either version 3 of the License, or (at your option) any later version.
//
// BOINC is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
// See the GNU Lesser General Public License for more details.
//
// You should have received a copy of the GNU Lesser General Public License
// along with BOINC. If not, see <http://www.gnu.org/licenses/>.
- server: change the following from per-host to per-(host, app version): - daily quota mechanism - reliable mechanism (accelerated retries) - "trusted" mechanism (adaptive replication) - scheduler: enforce host scale probation only for apps with host_scale_check set. - validator: do scale probation on invalid results (need this in addition to error and timeout cases) - feeder: update app version scales every 10 min, not 10 sec - back-end apps: support --foo as well as -foo for options Notes: - If you have, say, cuda, cuda23 and cuda_fermi plan classes, a host will have separate quotas for each one. That means it could error out on 100 jobs for cuda_fermi, and when its quota goes to zero, error out on 100 jobs for cuda23, etc. This is intentional; there may be cases where one version works but not the others. - host.error_rate and host.max_results_day are deprecated TODO: - the values in the app table for limits on jobs in progress etc. should override rather than config.xml. Implementation notes: scheduler: process_request(): read all host_app_versions for host at start; Compute "reliable" and "trusted" for each one. write modified records at end get_app_version(): add "reliable_only" arg; if set, use only reliable versions skip over-quota versions Multi-pass scheduling: if have at least one reliable version, do a pass for jobs that need reliable, and use only reliable versions. Then clear best_app_versions cache. Score-based scheduling: for need-reliable jobs, it will pick the fastest version, then give a score bonus if that version happens to be reliable. When get back a successful result from client: increase daily quota When get back an error result from client: impose scale probation decrease daily quota if not aborted Validator: when handling a WU, create a vector of HOST_APP_VERSION parallel to vector of RESULT. Pass it to assign_credit_set(). Make copies of originals so we can update only modified ones update HOST_APP_VERSION error rates Transitioner: decrease quota on timeout svn path=/trunk/boinc/; revision=21181
2010-04-15 03:13:56 +00:00
// wu_check
// look for results with missing input files
- server: change the following from per-host to per-(host, app version): - daily quota mechanism - reliable mechanism (accelerated retries) - "trusted" mechanism (adaptive replication) - scheduler: enforce host scale probation only for apps with host_scale_check set. - validator: do scale probation on invalid results (need this in addition to error and timeout cases) - feeder: update app version scales every 10 min, not 10 sec - back-end apps: support --foo as well as -foo for options Notes: - If you have, say, cuda, cuda23 and cuda_fermi plan classes, a host will have separate quotas for each one. That means it could error out on 100 jobs for cuda_fermi, and when its quota goes to zero, error out on 100 jobs for cuda23, etc. This is intentional; there may be cases where one version works but not the others. - host.error_rate and host.max_results_day are deprecated TODO: - the values in the app table for limits on jobs in progress etc. should override rather than config.xml. Implementation notes: scheduler: process_request(): read all host_app_versions for host at start; Compute "reliable" and "trusted" for each one. write modified records at end get_app_version(): add "reliable_only" arg; if set, use only reliable versions skip over-quota versions Multi-pass scheduling: if have at least one reliable version, do a pass for jobs that need reliable, and use only reliable versions. Then clear best_app_versions cache. Score-based scheduling: for need-reliable jobs, it will pick the fastest version, then give a score bonus if that version happens to be reliable. When get back a successful result from client: increase daily quota When get back an error result from client: impose scale probation decrease daily quota if not aborted Validator: when handling a WU, create a vector of HOST_APP_VERSION parallel to vector of RESULT. Pass it to assign_credit_set(). Make copies of originals so we can update only modified ones update HOST_APP_VERSION error rates Transitioner: decrease quota on timeout svn path=/trunk/boinc/; revision=21181
2010-04-15 03:13:56 +00:00
// --repair change them to server_state OVER, outcome COULDNT_SEND
//
// NOTE 1: this assumes that jobs have a single input file.
// NOTE 2: should rewrite to enumerate WUs, not results
#include "config.h"
#include <cstdio>
#include <cstdlib>
#include <cstring>
#include <string>
#include "boinc_db.h"
#include "svn_version.h"
#include "parse.h"
#include "util.h"
#include "error_numbers.h"
#include "sched_config.h"
#include "sched_util.h"
bool repair = false;
// wu_check
// See whether input files that should be present, are
// get the path a WU's input file
//
int get_file_path(WORKUNIT& wu, char* path) {
char buf[256];
bool flag;
flag = parse_str(wu.xml_doc, "<name>", buf, sizeof(buf));
if (!flag) return ERR_XML_PARSE;
dir_hier_path(buf, config.download_dir, config.uldl_dir_fanout, path);
return 0;
}
int handle_result(DB_RESULT& result) {
DB_WORKUNIT wu;
int retval;
char path[256];
char buf[256];
FILE* f;
retval = wu.lookup_id(result.workunitid);
if (retval) {
printf(
"ERROR: can't find WU %d for result %d\n",
result.workunitid, result.id
);
return 1;
}
get_file_path(wu, path);
f = fopen(path, "r");
if (f) {
fclose(f);
} else {
printf("no file %s for result %d\n",
path, result.id
);
if (repair) {
if (result.server_state == RESULT_SERVER_STATE_UNSENT) {
result.server_state = RESULT_SERVER_STATE_OVER;
result.outcome = RESULT_OUTCOME_COULDNT_SEND;
sprintf(
buf,"server_state=%d, outcome=%d",
result.server_state, result.outcome
);
retval = result.update_field(buf);
if (retval) {
printf(
"ERROR: can't update result %d\n",
result.id
);
return 1;
}
}
}
return 1;
}
return 0;
}
void usage(char *name) {
fprintf(stderr,
"Looks for results with missing input files\n\n"
"Usage: %s [OPTION]\n\n"
"Options:\n"
- server: change the following from per-host to per-(host, app version): - daily quota mechanism - reliable mechanism (accelerated retries) - "trusted" mechanism (adaptive replication) - scheduler: enforce host scale probation only for apps with host_scale_check set. - validator: do scale probation on invalid results (need this in addition to error and timeout cases) - feeder: update app version scales every 10 min, not 10 sec - back-end apps: support --foo as well as -foo for options Notes: - If you have, say, cuda, cuda23 and cuda_fermi plan classes, a host will have separate quotas for each one. That means it could error out on 100 jobs for cuda_fermi, and when its quota goes to zero, error out on 100 jobs for cuda23, etc. This is intentional; there may be cases where one version works but not the others. - host.error_rate and host.max_results_day are deprecated TODO: - the values in the app table for limits on jobs in progress etc. should override rather than config.xml. Implementation notes: scheduler: process_request(): read all host_app_versions for host at start; Compute "reliable" and "trusted" for each one. write modified records at end get_app_version(): add "reliable_only" arg; if set, use only reliable versions skip over-quota versions Multi-pass scheduling: if have at least one reliable version, do a pass for jobs that need reliable, and use only reliable versions. Then clear best_app_versions cache. Score-based scheduling: for need-reliable jobs, it will pick the fastest version, then give a score bonus if that version happens to be reliable. When get back a successful result from client: increase daily quota When get back an error result from client: impose scale probation decrease daily quota if not aborted Validator: when handling a WU, create a vector of HOST_APP_VERSION parallel to vector of RESULT. Pass it to assign_credit_set(). Make copies of originals so we can update only modified ones update HOST_APP_VERSION error rates Transitioner: decrease quota on timeout svn path=/trunk/boinc/; revision=21181
2010-04-15 03:13:56 +00:00
" [ --repair ] change them to server_state OVER,\n"
" outcome COULDNT_SEND\n"
- server: change the following from per-host to per-(host, app version): - daily quota mechanism - reliable mechanism (accelerated retries) - "trusted" mechanism (adaptive replication) - scheduler: enforce host scale probation only for apps with host_scale_check set. - validator: do scale probation on invalid results (need this in addition to error and timeout cases) - feeder: update app version scales every 10 min, not 10 sec - back-end apps: support --foo as well as -foo for options Notes: - If you have, say, cuda, cuda23 and cuda_fermi plan classes, a host will have separate quotas for each one. That means it could error out on 100 jobs for cuda_fermi, and when its quota goes to zero, error out on 100 jobs for cuda23, etc. This is intentional; there may be cases where one version works but not the others. - host.error_rate and host.max_results_day are deprecated TODO: - the values in the app table for limits on jobs in progress etc. should override rather than config.xml. Implementation notes: scheduler: process_request(): read all host_app_versions for host at start; Compute "reliable" and "trusted" for each one. write modified records at end get_app_version(): add "reliable_only" arg; if set, use only reliable versions skip over-quota versions Multi-pass scheduling: if have at least one reliable version, do a pass for jobs that need reliable, and use only reliable versions. Then clear best_app_versions cache. Score-based scheduling: for need-reliable jobs, it will pick the fastest version, then give a score bonus if that version happens to be reliable. When get back a successful result from client: increase daily quota When get back an error result from client: impose scale probation decrease daily quota if not aborted Validator: when handling a WU, create a vector of HOST_APP_VERSION parallel to vector of RESULT. Pass it to assign_credit_set(). Make copies of originals so we can update only modified ones update HOST_APP_VERSION error rates Transitioner: decrease quota on timeout svn path=/trunk/boinc/; revision=21181
2010-04-15 03:13:56 +00:00
" [ -h | --help ] Shows this help text\n"
" [ -v | --version ] Shows version information\n",
name
);
}
int main(int argc, char** argv) {
DB_RESULT result;
char clause[256];
int retval, n, nerr;
for(int c = 1; c < argc; c++) {
std::string option(argv[c]);
- server: change the following from per-host to per-(host, app version): - daily quota mechanism - reliable mechanism (accelerated retries) - "trusted" mechanism (adaptive replication) - scheduler: enforce host scale probation only for apps with host_scale_check set. - validator: do scale probation on invalid results (need this in addition to error and timeout cases) - feeder: update app version scales every 10 min, not 10 sec - back-end apps: support --foo as well as -foo for options Notes: - If you have, say, cuda, cuda23 and cuda_fermi plan classes, a host will have separate quotas for each one. That means it could error out on 100 jobs for cuda_fermi, and when its quota goes to zero, error out on 100 jobs for cuda23, etc. This is intentional; there may be cases where one version works but not the others. - host.error_rate and host.max_results_day are deprecated TODO: - the values in the app table for limits on jobs in progress etc. should override rather than config.xml. Implementation notes: scheduler: process_request(): read all host_app_versions for host at start; Compute "reliable" and "trusted" for each one. write modified records at end get_app_version(): add "reliable_only" arg; if set, use only reliable versions skip over-quota versions Multi-pass scheduling: if have at least one reliable version, do a pass for jobs that need reliable, and use only reliable versions. Then clear best_app_versions cache. Score-based scheduling: for need-reliable jobs, it will pick the fastest version, then give a score bonus if that version happens to be reliable. When get back a successful result from client: increase daily quota When get back an error result from client: impose scale probation decrease daily quota if not aborted Validator: when handling a WU, create a vector of HOST_APP_VERSION parallel to vector of RESULT. Pass it to assign_credit_set(). Make copies of originals so we can update only modified ones update HOST_APP_VERSION error rates Transitioner: decrease quota on timeout svn path=/trunk/boinc/; revision=21181
2010-04-15 03:13:56 +00:00
if(option == "-h" || option == "--help") {
usage(argv[0]);
exit(0);
- server: change the following from per-host to per-(host, app version): - daily quota mechanism - reliable mechanism (accelerated retries) - "trusted" mechanism (adaptive replication) - scheduler: enforce host scale probation only for apps with host_scale_check set. - validator: do scale probation on invalid results (need this in addition to error and timeout cases) - feeder: update app version scales every 10 min, not 10 sec - back-end apps: support --foo as well as -foo for options Notes: - If you have, say, cuda, cuda23 and cuda_fermi plan classes, a host will have separate quotas for each one. That means it could error out on 100 jobs for cuda_fermi, and when its quota goes to zero, error out on 100 jobs for cuda23, etc. This is intentional; there may be cases where one version works but not the others. - host.error_rate and host.max_results_day are deprecated TODO: - the values in the app table for limits on jobs in progress etc. should override rather than config.xml. Implementation notes: scheduler: process_request(): read all host_app_versions for host at start; Compute "reliable" and "trusted" for each one. write modified records at end get_app_version(): add "reliable_only" arg; if set, use only reliable versions skip over-quota versions Multi-pass scheduling: if have at least one reliable version, do a pass for jobs that need reliable, and use only reliable versions. Then clear best_app_versions cache. Score-based scheduling: for need-reliable jobs, it will pick the fastest version, then give a score bonus if that version happens to be reliable. When get back a successful result from client: increase daily quota When get back an error result from client: impose scale probation decrease daily quota if not aborted Validator: when handling a WU, create a vector of HOST_APP_VERSION parallel to vector of RESULT. Pass it to assign_credit_set(). Make copies of originals so we can update only modified ones update HOST_APP_VERSION error rates Transitioner: decrease quota on timeout svn path=/trunk/boinc/; revision=21181
2010-04-15 03:13:56 +00:00
} else if(option == "-v" || option == "--version") {
printf("%s\n", SVN_VERSION);
exit(0);
- server: change the following from per-host to per-(host, app version): - daily quota mechanism - reliable mechanism (accelerated retries) - "trusted" mechanism (adaptive replication) - scheduler: enforce host scale probation only for apps with host_scale_check set. - validator: do scale probation on invalid results (need this in addition to error and timeout cases) - feeder: update app version scales every 10 min, not 10 sec - back-end apps: support --foo as well as -foo for options Notes: - If you have, say, cuda, cuda23 and cuda_fermi plan classes, a host will have separate quotas for each one. That means it could error out on 100 jobs for cuda_fermi, and when its quota goes to zero, error out on 100 jobs for cuda23, etc. This is intentional; there may be cases where one version works but not the others. - host.error_rate and host.max_results_day are deprecated TODO: - the values in the app table for limits on jobs in progress etc. should override rather than config.xml. Implementation notes: scheduler: process_request(): read all host_app_versions for host at start; Compute "reliable" and "trusted" for each one. write modified records at end get_app_version(): add "reliable_only" arg; if set, use only reliable versions skip over-quota versions Multi-pass scheduling: if have at least one reliable version, do a pass for jobs that need reliable, and use only reliable versions. Then clear best_app_versions cache. Score-based scheduling: for need-reliable jobs, it will pick the fastest version, then give a score bonus if that version happens to be reliable. When get back a successful result from client: increase daily quota When get back an error result from client: impose scale probation decrease daily quota if not aborted Validator: when handling a WU, create a vector of HOST_APP_VERSION parallel to vector of RESULT. Pass it to assign_credit_set(). Make copies of originals so we can update only modified ones update HOST_APP_VERSION error rates Transitioner: decrease quota on timeout svn path=/trunk/boinc/; revision=21181
2010-04-15 03:13:56 +00:00
} else if (option == "--repair") {
repair = true;
} else {
fprintf(stderr, "unknown command line argument: %s\n\n", argv[c]);
usage(argv[0]);
exit(1);
}
}
retval = config.parse_file();
if (retval) exit(1);
retval = boinc_db.open(config.db_name, config.db_host, config.db_user, config.db_passwd);
if (retval) {
printf("boinc_db.open: %d\n", retval);
exit(1);
}
n = nerr = 0;
printf("Unsent results:\n");
sprintf(clause, "where server_state=%d", RESULT_SERVER_STATE_UNSENT);
while (!result.enumerate(clause)) {
retval = handle_result(result);
n++;
if (retval) nerr++;
}
printf("%d out of %d errors\n", nerr, n);
n = nerr = 0;
printf("In progress results:\n");
sprintf(clause, "where server_state=%d", RESULT_SERVER_STATE_IN_PROGRESS);
while (!result.enumerate(clause)) {
retval = handle_result(result);
n++;
if (retval) nerr++;
}
printf("%d out of %d errors\n", nerr, n);
}
const char *BOINC_RCSID_8f4e399992 = "$Id$";