~ubuntu-branches/ubuntu/utopic/gridengine/utopic

It contains information about the cluster (hosts, queues, load values, consumables, etc.), about the jobs running in the cluster and about sharetree configuration and usage.

All information is time related, events are dumped to the reporting file in a configurable interval.

It allows to monitor a "real time" status of the cluster as well as historical analysis.

.\"

.SH FORMAT

The reporting file is an ASCII file.

Each line contains one record, and the fields of a record are separated by a delimiter (:).

The reporting file contains records of different type. Each record type has a specific record structure.

.PP

The first two fields are common to all reporting records:

.IP "\fBtime\fP"

Time (GMT unix timestamp) when the record was created.

.IP "\fBrecord type\fP"

Type of the accounting record.

The different types of records and their structure are described in the following text.

.SS new_job

The new_job record is written whenever a new job enters the system (usually by a submitting command). It has the following fields:

.IP "\fBsubmission_time\fP"

Time (GMT unix time stamp) when the job was submitted.

.IP "\fBjob_number\fP"

The job number.

.IP "\fBtask_number\fP"

The array task id. Always has the value -1 for new_job records (as we don't have array tasks yet).

.IP "\fBpe_taskid\fP"

The task id of parallel tasks. Always has the value "none" for new_job records.

.IP "\fBjob_name\fP"

The job name (from -N submission option)

.IP "\fBowner\fP"

The job owner.

.IP "\fBgroup\fP"

The unix group of the job owner.

.IP "\fBproject\fP"

The project the job is running in.

.IP "\fBdepartment\fP"

The department the job owner is in.

.IP "\fBaccount\fP"

The account string specified for the job (from -A submission option).

.IP "\fBpriority\fP"

The job priority (from -p submission option).

.SS job_log

The job_log record is written whenever a job, an array task or a pe tasks is changing status. A status change can be the transition from pending to running, but can also be triggered by user actions like suspension of a job.

It has the following fields:

.IP "\fBevent_time\fP"

Time (GMT unix time stamp) when the event was generated.

.IP "\fBevent\fP"

A one word description of the event.

.IP "\fBjob_number\fP"

The job number.

.IP "\fBtask_number\fP"

The array task id. Always has the value -1 for new_job records (as we don't have array tasks yet).

.IP "\fBpe_taskid\fP"

The task id of parallel tasks. Always has the value "none" for new_job records.

.IP "\fBstate\fP"

The state of the job after the event was processed.

.IP "\fBuser\fP"

The user who initiated the event (or special usernames "qmaster", "scheduler"

and "execd" for actions of the system itself like scheduling jobs, executing jobs etc.).

.IP "\fBhost\fP"

100

The host from which the action was initiated (e.g. the submit host, the qmaster host, etc.).

101

.IP "\fBstate_time\fP"

102

Reserved field for later use.

103

.IP "\fBsubmission_time\fP"

104

Time (GMT unix time stamp) when the job was submitted.

105

.IP "\fBjob_name\fP"

106

The job name (from -N submission option)

107

.IP "\fBowner\fP"

108

The job owner.

109

.IP "\fBgroup\fP"

110

The unix group of the job owner.

111

.IP "\fBproject\fP"

112

The project the job is running in.

113

.IP "\fBdepartment\fP"

114

The department the job owner is in.

115

.IP "\fBaccount\fP"

116

The account string specified for the job (from -A submission option).

117

.IP "\fBpriority\fP"

118

The job priority (from -p submission option).

119

.IP "\fBmessage\fP"

120

A message describing the reported action.

121

.SS acct

122

Records of type acct are accounting records. Normally, they are written whenever a job, a task of an array job,

123

or the task of a parallel job terminates. However, for long running jobs an intermediate acct record is created once a

124

day after a midnight. This results in multiple accounting records for a particular job and allows for a fine-grained

125

resource usage monitoring over time.

126

Accounting records comprise the following fields:

127

.IP "\fBqname\fP"

128

Name of the cluster queue in which the job has run.

129

.IP "\fBhostname\fP"

130

Name of the execution host.

131

.IP "\fBgroup\fP"

132

The effective group id of the job owner when executing the job.

133

.IP "\fBowner\fP"

134

Owner of the xxQS_NAMExx job.

135

.IP "\fBjob_name\fP"

136

Job name.

137

.IP "\fBjob_number\fP"

138

Job identifier - job number.

139

.IP "\fBaccount\fP"

140

An account string as specified by the

141

.M qsub 1

142

143

.M qalter 1

144

\fB\-A\fP option.

145

.IP "\fBpriority\fP"

146

Priority value assigned to the job corresponding to the \fBpriority\fP

147

parameter in the queue configuration (see

148

.M queue_conf 5 ).

149

.IP "\fBsubmission_time\fP"

150

Submission time (GMT unix time stamp).

151

.IP "\fBstart_time\fP"

152

Start time (GMT unix time stamp).

153

.IP "\fBend_time\fP"

154

End time (GMT unix time stamp).

155

.IP "\fBfailed\fP"

156

Indicates the problem which occurred in case a job could not be started on

157

the execution host (e.g. because the owner of the job did not have a valid

158

account on that machine). If xxQS_NAMExx tries to start a job multiple times,

159

this may lead to multiple entries in the accounting file corresponding to

160

the same job ID.

161

.IP "\fBexit_status\fP"

162

Exit status of the job script (or xxQS_NAMExx specific status in case

163

of certain error conditions).

164

.IP "\fBru_wallclock\fP"

165

Difference between \fBend_time\fP and \fBstart_time\fP (see above).

166

.PP

167

The remainder of the accounting entries follows the contents of the

168

standard UNIX rusage structure as described in

169

.M getrusage 2 .

170

Depending on the operating system where the job was executed some of the

171

fields may be 0. The following entries are provided:

172

.PP

173

.nf

174

.RS

175

.B ru_utime

176

.B ru_stime

177

.B ru_maxrss

178

.B ru_ixrss

179

.B ru_ismrss

180

.B ru_idrss

181

.B ru_isrss

182

.B ru_minflt

183

.B ru_majflt

184

.B ru_nswap

185

.B ru_inblock

186

.B ru_oublock

187

.B ru_msgsnd

188

.B ru_msgrcv

189

.B ru_nsignals

190

.B ru_nvcsw

191

.B ru_nivcsw

192

.RE

193

.fi

194

.PP

195

196

.IP "\fBproject\fP"

197

The project which was assigned to the job.

198

.IP "\fBdepartment\fP"

199

The department which was assigned to the job.

200

.IP "\fBgranted_pe\fP"

201

The parallel environment which was selected for that job.

202

.IP "\fBslots\fP"

203

The number of slots which were dispatched to the job by the scheduler.

204

.IP "\fBtask_number\fP"

205

Array job task index number.

206

.IP "\fBcpu\fP"

207

The cpu time usage in seconds.

208

.IP "\fBmem\fP"

209

The integral memory usage in Gbytes seconds.

210

.IP "\fBio\fP"

211

The amount of data transferred in input/output operations.

212

.IP "\fBcategory\fP"

213

A string specifying the job category.

214

.IP "\fBiow\fP"

215

The io wait time in seconds.

216

.IP "\fBpe_taskid\fP"

217

If this identifier is set the task was part of a parallel job and was

218

passed to xxQS_NAMExx via the qrsh -inherit interface.

219

.IP "\fBmaxvmem\fP"

220

The maximum vmem size in bytes.

221

.IP "\fBarid\fP"

222

Advance reservation identifier. If the job used resources of an advance

223

reservation then this field contains a positive integer identifier otherwise the

224

value is "\fB0\fP" .

225

.SS queue

226

Records of type queue contain state information for queues (queue instances).

227

A queue record has the following fields:

228

.IP "\fBqname\fP"

229

The cluster queue name.

230

.IP "\fBhostname\fP"

231

The hostname of a specific queue instance.

232

.IP "\fBreport_time\fP"

233

The time (GMT unix time stamp) when a state change was triggered.

234

.IP "\fBstate\fP"

235

The new queue state.

236

.SS queue_consumable

237

A queue_consumable record contains information about queue consumable values in addition to queue state information:

238

.IP "\fBqname\fP"

239

The cluster queue name.

240

.IP "\fBhostname\fP"

241

The hostname of a specific queue instance.

242

.IP "\fBreport_time\fP"

243

The time (GMT unix time stamp) when a state change was triggered.

244

.IP "\fBstate\fP"

245

The new queue state.

246

.IP "\fBconsumables\fP"

247

Description of consumable values. Information about multiple consumables is separated by space.

248

A consumable description has the format <name>=<actual_value>=<configured value>.

249

.SS host

250

A host record contains information about hosts and host load values.

251

It contains the following information:

252

.IP "\fBhostname\fP"

253

The name of the host.

254

.IP "\fBreport_time\fP"

255

The time (GMT unix time stamp) when the reported information was generated.

256

.IP "\fBstate\fP"

257

The new host state.

258

Currently, xxQS_NAMExx doesn't track a host state, the field is reserved for

259

future use. Always contains the value X.

260

.IP "\fBload values\fP"

261

Description of load values. Information about multiple load values is separated by space.

262

A load value description has the format <name>=<actual_value>.

263

.\"

264

.SS host_consumable

265

A host_consumable record contains information about hosts and host consumables.

266

Host consumables can for example be licenses.

267

It contains the following information:

268

.IP "\fBhostname\fP"

269

The name of the host.

270

.IP "\fBreport_time\fP"

271

The time (GMT unix time stamp) when the reported information was generated.

272

.IP "\fBstate\fP"

273

The new host state.

274

Currently, xxQS_NAMExx doesn't track a host state, the field is reserved for

275

future use. Always contains the value X.

276

.IP "\fBconsumables\fP"

277

Description of consumable values. Information about multiple consumables is separated by space.

278

A consumable description has the format <name>=<actual_value>=<configured value>.

279

.SS sharelog

280

The xxQS_NAMExx qmaster can dump information about sharetree configuration and use to the reporting file.

281

The parameter \fIsharelog\fP sets an interval in which sharetree information will be dumped.

282

It is set in the format HH:MM:SS. A value of 00:00:00 configures qmaster not to

283

dump sharetree information. Intervals of several minutes up to hours are sensible values for this parameter.

284

The record contains the following fields

285

.IP "\fBcurrent time\fP"

286

The present time

287

.IP "\fBusage time\fP"

288

The time used so far

289

.IP "\fBnode name\fP"

290

The node name

291

.IP "\fBuser name\fP"

292

The user name

293

.IP "\fBproject name\fP"

294

The project name

295

.IP "\fBshares \fP"

296

The total shares

297

.IP "\fBjob count \fP"

298

The job count

299

.IP "\fBlevel \fP"

300

The percentage of shares used

301

.IP "\fBtotal \fP"

302

The adjusted percentage of shares used

303

.IP "\fBlong target share \fP"

304

The long target percentage of resource shares used

305

.IP "\fBshort target share \fP"

306

The short target percentage of resource shares used

307

.IP "\fBactual share \fP"

308

The actual percentage of resource shares used

309

.IP "\fBusage \fP"

310

The combined shares used

311

.IP "\fBcpu \fP"

312

The cpu used

313

.IP "\fBmem \fP"

314

The memory used

315

.IP "\fBio \fP"

316

The IO used

317

.IP "\fBlong target cpu \fP"

318

The long target cpu used

319

.IP "\fBlong target mem \fP"

320

The long target memory used

321

.IP "\fBlong target io \fP"

322

The long target IO used

323

.\"

324

.SS new_ar

325

A new_ar record contains information about advance reservation objects. Entries of this

326

type will be added if an advance reservation is created.

327

It contains the following information:

328

.IP "\fBsubmission_time\fP"

329

The time (GMT unix time stamp) when the advance reservation was created.

330

.IP "\fBar_number\fP"

331

The advance reservation number identifying the reservation.

332

.IP "\fBar_owner\fP"

333

The owner of the advance reservation.

334

.\"

335

.SS ar_attribute

336

The ar_attribute record is written whenever a new advance reservation was added or the

337

attribute of an existing advance reservation has changed. It has following fields.

338

.IP "\fBevent_time\fP"

339

The time (GMT unix time stamp) when the event was generated.

340

.IP "\fBsubmission_time\fP"

341

The time (GMT unix time stamp) when the advance reservation was created.

342

.IP "\fBar_number\fP"

343

The advance reservation number identifying the reservation.

344

.IP "\fBar_name\fP"

345

Name of the advance reservation.

346

.IP "\fBar_account\fP"

347

An account string which was specified during the creation of the advance reservation.

348

.IP "\fBar_start_time\fP"

349

Start time.

350

.IP "\fBar_end_time\fP"

351

End time.

352

.IP "\fBar_granted_pe\fP"

353

The parallel environment which was selected for an advance reservation.

354

.IP "\fBar_granted_resources\fP"

355

The granted resources which were selected for an advance reservation.

356

.\"

357

.SS ar_log

358

The ar_log record is written whenever a advance reservation is changing status. A status

359

change can be from pending to active, but can also be triggered by system events like host

360

outage. It has following fields.

361

.IP "\fBar_state_change_time\fP"

362

The time (GMT unix time stamp) when the event occurred which caused a state change.

363

.IP "\fBsubmission_time\fP"

364

The time (GMT unix time stamp) when the advance reservation was created.

365

.IP "\fBar_number\fP"

366

The advance reservation number identifying the reservation.

367

.IP "\fBar_state\fP"

368

The new state.

369

.IP "\fBar_event\fP"

370

An event id identifying the event which caused the state change.

371

.IP "\fBar_message\fP"

372

A message describing the event which caused the state change.

373

.\"

374

.SS ar_acct

375

The ar_acct records are accounting records which are written for every queue instance

376

whenever a advance reservation terminates. Advance reservation accounting records comprise

377

following fields.

378

.IP "\fBar_termination_time\fP"

379

The time (GMT unix time stamp) when the advance reservation terminated.

380

.IP "\fBsubmission_time\fP"

381

The time (GMT unix time stamp) when the advance reservation was created.

382

.IP "\fBar_number\fP"

383

The advance reservation number identifying the reservation.

384

.IP "\fBar_qname\fP"

385

Cluster queue name which the advance reservation reserved.

386

.IP "\fBar_hostname\fP"

387

The name of the execution host.

388

.IP "\fBar_slots\fP"

389

The number of slots which were reserved.

390

.\"

391

.\"

392

.SH "SEE ALSO"

393

.M sge_conf 5 .

394

.M host_conf 5 .

395

.\"

396

.SH "COPYRIGHT"

397

See

398

.M xxqs_name_sxx_intro 1

399

for a full statement of rights and permissions.

Older »