Ottomata has uploaded a new change for review.

  https://gerrit.wikimedia.org/r/231785

Change subject: Split webrequest camus import job into multiple jobs for 
different size topics
......................................................................

Split webrequest camus import job into multiple jobs for different size topics

Change-Id: Ideb4ed885360e9562734526dca1f524506280de8
---
M manifests/role/analytics/refinery.pp
1 file changed, 22 insertions(+), 4 deletions(-)


  git pull ssh://gerrit.wikimedia.org:29418/operations/puppet 
refs/changes/85/231785/1

diff --git a/manifests/role/analytics/refinery.pp 
b/manifests/role/analytics/refinery.pp
index b449a39..8f4f5e7 100644
--- a/manifests/role/analytics/refinery.pp
+++ b/manifests/role/analytics/refinery.pp
@@ -54,13 +54,31 @@
 class role::analytics::refinery::camus {
     require role::analytics::refinery
 
-    $camus_webrequest_properties = 
"${::role::analytics::refinery::path}/camus/camus.webrequest.properties"
-    $camus_webrequest_log_file   = 
"${::role::analytics::refinery::log_dir}/camus-webrequest.log"
-    cron { 'refinery-camus-webrequest-import':
-        command => "${::role::analytics::refinery::path}/bin/camus --job-name 
refinery-camus-webrequest-import ${camus_webrequest_properties} >> 
${camus_webrequest_log_file} 2>&1",
+    # TODO: Move these Camus .properties files to puppet erb templates.
+
+    # Split up high and low volume webrequest Camus imports into multiple jobs.
+    # This will help keep large volume imports from causing lower volume ones
+    # to lag.
+
+    # webrequest_text
+    cron { 'refinery-camus-webrequest_text-import':
+        command => "${::role::analytics::refinery::path}/bin/camus --job-name 
refinery-camus-webrequest_text-import 
${::role::analytics::refinery::path}/camus/camus.webrequest_text.properties >> 
${::role::analytics::refinery::log_dir}/camus-webrequest_text.log 2>&1",
         user    => 'hdfs',  # we might want to use a different user for this, 
not sure.
         minute  => '*/10',
     }
+    # webrequest_upload
+    cron { 'refinery-camus-webrequest_upload-import':
+        command => "${::role::analytics::refinery::path}/bin/camus --job-name 
refinery-camus-webrequest_upload-import 
${::role::analytics::refinery::path}/camus/camus.webrequest_upload.properties 
>> ${::role::analytics::refinery::log_dir}/camus-webrequest_upload.log 2>&1",
+        user    => 'hdfs',  # we might want to use a different user for this, 
not sure.
+        minute  => '*/10',
+    }
+    # webrequest_text
+    cron { 'refinery-camus-webrequest-small-import':
+        command => "${::role::analytics::refinery::path}/bin/camus --job-name 
refinery-camus-webrequest-small-import 
${::role::analytics::refinery::path}/camus/camus.webrequest-small.properties >> 
${::role::analytics::refinery::log_dir}/camus-webrequest-small.log 2>&1",
+        user    => 'hdfs',  # we might want to use a different user for this, 
not sure.
+        minute  => '*/10',
+    }
+
 
     $camus_eventlogging_properties = 
"${::role::analytics::refinery::path}/camus/camus.eventlogging.properties"
     $camus_eventlogging_log_file   = 
"${::role::analytics::refinery::log_dir}/camus-eventlogging.log"

-- 
To view, visit https://gerrit.wikimedia.org/r/231785
To unsubscribe, visit https://gerrit.wikimedia.org/r/settings

Gerrit-MessageType: newchange
Gerrit-Change-Id: Ideb4ed885360e9562734526dca1f524506280de8
Gerrit-PatchSet: 1
Gerrit-Project: operations/puppet
Gerrit-Branch: production
Gerrit-Owner: Ottomata <[email protected]>

_______________________________________________
MediaWiki-commits mailing list
[email protected]
https://lists.wikimedia.org/mailman/listinfo/mediawiki-commits

Reply via email to