diff --git a/content/post/2016-12.md b/content/post/2016-12.md index cae6f1155..3c865e9d4 100644 --- a/content/post/2016-12.md +++ b/content/post/2016-12.md @@ -684,3 +684,16 @@ $ exit ``` $ [dspace]/bin/dspace metadata-export -a -f /tmp/iita.csv -i 10568/68616 ``` + +## 2016-12-28 + +- We've been getting two alerts per day about CPU usage on the new server from Linode +- These are caused by the batch jobs for Solr etc that run in the early morning hours +- The Linode default is to alert at 90% CPU usage for two hours, but I see the old server was at 150%, so maybe we just need to adjust it +- Speaking of the old server (linode01), I think we can decommission it now +- I checked the S3 logs on the new server (linode18) to make sure the backups have been running and everything looks good +- In other news, I was looking at the Munin graphs for PostgreSQL on the new server and it looks slightly worrying: + +![munin postgres stats](2016/12/postgres_size_ALL-week.png) + +- I will have to check later why the size keeps increasing diff --git a/public/2015-11/index.html b/public/2015-11/index.html index dae7e9ed3..6b1567f66 100644 --- a/public/2015-11/index.html +++ b/public/2015-11/index.html @@ -90,7 +90,7 @@ $ psql -c 'SELECT * from pg_stat_activity;' | grep idle | grep -c cgspac - + @@ -119,6 +119,14 @@ $ psql -c 'SELECT * from pg_stat_activity;' | grep idle | grep -c cgspac Home + CGSpace Notes + + Notes + + Posts + + Tags + diff --git a/public/2015-12/index.html b/public/2015-12/index.html index 5ca3d2c95..e6605b2bc 100644 --- a/public/2015-12/index.html +++ b/public/2015-12/index.html @@ -93,7 +93,7 @@ Replace lzop with xz in log compression cron jobs on DSpace Test—it uses less - + @@ -122,6 +122,14 @@ Replace lzop with xz in log compression cron jobs on DSpace Test—it uses less Home + CGSpace Notes + + Notes + + Posts + + Tags + diff --git a/public/2016-01/index.html b/public/2016-01/index.html index 7f3b110cd..76b2f9be8 100644 --- a/public/2016-01/index.html +++ b/public/2016-01/index.html @@ -78,7 +78,7 @@ Update GitHub wiki for documentation of maintenance tasks. - + @@ -107,6 +107,14 @@ Update GitHub wiki for documentation of maintenance tasks. Home + CGSpace Notes + + Notes + + Posts + + Tags + diff --git a/public/2016-02/index.html b/public/2016-02/index.html index c35e532a9..853740b81 100644 --- a/public/2016-02/index.html +++ b/public/2016-02/index.html @@ -99,7 +99,7 @@ Also, lots of things like “COTE D`LVOIRE” and “COTE D IVOIRE&r - + @@ -128,6 +128,14 @@ Also, lots of things like “COTE D`LVOIRE” and “COTE D IVOIRE&r Home + CGSpace Notes + + Notes + + Posts + + Tags + diff --git a/public/2016-03/index.html b/public/2016-03/index.html index 1be8e6bad..fed90dc5c 100644 --- a/public/2016-03/index.html +++ b/public/2016-03/index.html @@ -78,7 +78,7 @@ Reinstall my local (Mac OS X) DSpace stack with Tomcat 7, PostgreSQL 9.3, and Ja - + @@ -107,6 +107,14 @@ Reinstall my local (Mac OS X) DSpace stack with Tomcat 7, PostgreSQL 9.3, and Ja Home + CGSpace Notes + + Notes + + Posts + + Tags + diff --git a/public/2016-04/index.html b/public/2016-04/index.html index 84be8b89b..f4042f3b2 100644 --- a/public/2016-04/index.html +++ b/public/2016-04/index.html @@ -84,7 +84,7 @@ Also, I noticed the checker log has some errors we should pay attention to: - + @@ -113,6 +113,14 @@ Also, I noticed the checker log has some errors we should pay attention to: Home + CGSpace Notes + + Notes + + Posts + + Tags + diff --git a/public/2016-05/index.html b/public/2016-05/index.html index 4761a9600..18e9a59ac 100644 --- a/public/2016-05/index.html +++ b/public/2016-05/index.html @@ -90,7 +90,7 @@ There are 3,000 IPs accessing the REST API in a 24-hour period! - + @@ -119,6 +119,14 @@ There are 3,000 IPs accessing the REST API in a 24-hour period! Home + CGSpace Notes + + Notes + + Posts + + Tags + diff --git a/public/2016-06/index.html b/public/2016-06/index.html index 831ee9346..04cc06ccd 100644 --- a/public/2016-06/index.html +++ b/public/2016-06/index.html @@ -87,7 +87,7 @@ Working on second phase of metadata migration, looks like this will work for mov - + @@ -116,6 +116,14 @@ Working on second phase of metadata migration, looks like this will work for mov Home + CGSpace Notes + + Notes + + Posts + + Tags + diff --git a/public/2016-07/index.html b/public/2016-07/index.html index 61079dd6a..d0786b6f3 100644 --- a/public/2016-07/index.html +++ b/public/2016-07/index.html @@ -111,7 +111,7 @@ In this case the select query was showing 95 results before the update - + @@ -140,6 +140,14 @@ In this case the select query was showing 95 results before the update Home + CGSpace Notes + + Notes + + Posts + + Tags + diff --git a/public/2016-08/index.html b/public/2016-08/index.html index 90ec6e09d..7d6033d16 100644 --- a/public/2016-08/index.html +++ b/public/2016-08/index.html @@ -102,7 +102,7 @@ $ git rebase -i dspace-5.5 - + @@ -131,6 +131,14 @@ $ git rebase -i dspace-5.5 Home + CGSpace Notes + + Notes + + Posts + + Tags + diff --git a/public/2016-09/index.html b/public/2016-09/index.html index da15ab312..4b155adf3 100644 --- a/public/2016-09/index.html +++ b/public/2016-09/index.html @@ -90,7 +90,7 @@ $ ldapsearch -x -H ldaps://svcgroot2.cgiarad.org:3269/ -b "dc=cgiarad,dc=or - + @@ -119,6 +119,14 @@ $ ldapsearch -x -H ldaps://svcgroot2.cgiarad.org:3269/ -b "dc=cgiarad,dc=or Home + CGSpace Notes + + Notes + + Posts + + Tags + diff --git a/public/2016-10/index.html b/public/2016-10/index.html index 28c69bc86..2a6808452 100644 --- a/public/2016-10/index.html +++ b/public/2016-10/index.html @@ -54,7 +54,7 @@ - + @@ -83,6 +83,14 @@ Home + CGSpace Notes + + Notes + + Posts + + Tags + diff --git a/public/2016-11/index.html b/public/2016-11/index.html index 465e681ec..e7cfdacad 100644 --- a/public/2016-11/index.html +++ b/public/2016-11/index.html @@ -54,7 +54,7 @@ - + @@ -83,6 +83,14 @@ Home + CGSpace Notes + + Notes + + Posts + + Tags + diff --git a/public/2016-12/index.html b/public/2016-12/index.html index 1c00ff489..3ed76b40b 100644 --- a/public/2016-12/index.html +++ b/public/2016-12/index.html @@ -30,7 +30,7 @@ - + @@ -54,7 +54,7 @@ - + @@ -83,6 +83,14 @@ Home + CGSpace Notes + + Notes + + Posts + + Tags + @@ -851,6 +859,23 @@ $ exit 
$ [dspace]/bin/dspace metadata-export -a -f /tmp/iita.csv -i 10568/68616
+

2016-12-28

+ + + +

munin postgres stats

+ + + diff --git a/public/2016/12/postgres_size_ALL-week.png b/public/2016/12/postgres_size_ALL-week.png new file mode 100644 index 000000000..e2a6dabec Binary files /dev/null and b/public/2016/12/postgres_size_ALL-week.png differ diff --git a/public/index.html b/public/index.html index 0d70a08b8..b88549926 100644 --- a/public/index.html +++ b/public/index.html @@ -36,7 +36,7 @@ - + @@ -69,6 +69,14 @@ Home + CGSpace Notes + + Notes + + Posts + + Tags + @@ -344,7 +352,7 @@ dspacetest=# select text_value from metadatavalue where metadata_field_id=3 and Previous page - Next page + Next page diff --git a/public/index.xml b/public/index.xml index 94f7e800e..7e7d5b3c8 100644 --- a/public/index.xml +++ b/public/index.xml @@ -754,6 +754,23 @@ $ exit <pre><code>$ [dspace]/bin/dspace metadata-export -a -f /tmp/iita.csv -i 10568/68616 </code></pre> + +<h2 id="2016-12-28">2016-12-28</h2> + +<ul> +<li>We&rsquo;ve been getting two alerts per day about CPU usage on the new server from Linode</li> +<li>These are caused by the batch jobs for Solr etc that run in the early morning hours</li> +<li>The Linode default is to alert at 90% CPU usage for two hours, but I see the old server was at 150%, so maybe we just need to adjust it</li> +<li>Speaking of the old server (linode01), I think we can decommission it now</li> +<li>I checked the S3 logs on the new server (linode18) to make sure the backups have been running and everything looks good</li> +<li>In other news, I was looking at the Munin graphs for PostgreSQL on the new server and it looks slightly worrying:</li> +</ul> + +<p><img src="2016/12/postgres_size_ALL-week.png" alt="munin postgres stats" /></p> + +<ul> +<li>I will have to check later why the size keeps increasing</li> +</ul> diff --git a/public/page/2/index.html b/public/page/2/index.html index 2ff7da9d6..830b3648f 100644 --- a/public/page/2/index.html +++ b/public/page/2/index.html @@ -36,7 +36,7 @@ - + @@ -69,6 +69,14 @@ Home + CGSpace Notes + + Notes + + Posts + + Tags + @@ -206,7 +214,7 @@ @@ -344,7 +352,7 @@ dspacetest=# select text_value from metadatavalue where metadata_field_id=3 and Previous page - Next page + Next page diff --git a/public/post/index.xml b/public/post/index.xml index 1c795edc5..78795852c 100644 --- a/public/post/index.xml +++ b/public/post/index.xml @@ -1,9 +1,9 @@ - Post-rsses on CGSpace Notes + Posts on CGSpace Notes https://alanorth.github.io/cgspace-notes/post/index.xml - Recent content in Post-rsses on CGSpace Notes + Recent content in Posts on CGSpace Notes Hugo -- gohugo.io en-us Fri, 02 Dec 2016 10:43:00 +0300 @@ -754,6 +754,23 @@ $ exit <pre><code>$ [dspace]/bin/dspace metadata-export -a -f /tmp/iita.csv -i 10568/68616 </code></pre> + +<h2 id="2016-12-28">2016-12-28</h2> + +<ul> +<li>We&rsquo;ve been getting two alerts per day about CPU usage on the new server from Linode</li> +<li>These are caused by the batch jobs for Solr etc that run in the early morning hours</li> +<li>The Linode default is to alert at 90% CPU usage for two hours, but I see the old server was at 150%, so maybe we just need to adjust it</li> +<li>Speaking of the old server (linode01), I think we can decommission it now</li> +<li>I checked the S3 logs on the new server (linode18) to make sure the backups have been running and everything looks good</li> +<li>In other news, I was looking at the Munin graphs for PostgreSQL on the new server and it looks slightly worrying:</li> +</ul> + +<p><img src="2016/12/postgres_size_ALL-week.png" alt="munin postgres stats" /></p> + +<ul> +<li>I will have to check later why the size keeps increasing</li> +</ul> diff --git a/public/post/page/2/index.html b/public/post/page/2/index.html index 37a1688dd..a1a139507 100644 --- a/public/post/page/2/index.html +++ b/public/post/page/2/index.html @@ -36,7 +36,7 @@ - + @@ -69,6 +69,14 @@ Home + CGSpace Notes + + Notes + + Posts + + Tags + @@ -206,7 +214,7 @@ @@ -344,7 +352,7 @@ dspacetest=# select text_value from metadatavalue where metadata_field_id=3 and Previous page - Next page + Next page diff --git a/public/tags/notes/index.xml b/public/tags/notes/index.xml index 677caca60..9cee53775 100644 --- a/public/tags/notes/index.xml +++ b/public/tags/notes/index.xml @@ -1,9 +1,9 @@ - CGSpace Notes + Notes on CGSpace Notes https://alanorth.github.io/cgspace-notes/tags/notes/index.xml - Recent content on CGSpace Notes + Recent content in Notes on CGSpace Notes Hugo -- gohugo.io en-us @@ -753,6 +753,23 @@ $ exit <pre><code>$ [dspace]/bin/dspace metadata-export -a -f /tmp/iita.csv -i 10568/68616 </code></pre> + +<h2 id="2016-12-28">2016-12-28</h2> + +<ul> +<li>We&rsquo;ve been getting two alerts per day about CPU usage on the new server from Linode</li> +<li>These are caused by the batch jobs for Solr etc that run in the early morning hours</li> +<li>The Linode default is to alert at 90% CPU usage for two hours, but I see the old server was at 150%, so maybe we just need to adjust it</li> +<li>Speaking of the old server (linode01), I think we can decommission it now</li> +<li>I checked the S3 logs on the new server (linode18) to make sure the backups have been running and everything looks good</li> +<li>In other news, I was looking at the Munin graphs for PostgreSQL on the new server and it looks slightly worrying:</li> +</ul> + +<p><img src="2016/12/postgres_size_ALL-week.png" alt="munin postgres stats" /></p> + +<ul> +<li>I will have to check later why the size keeps increasing</li> +</ul> diff --git a/public/tags/notes/page/2/index.html b/public/tags/notes/page/2/index.html index f9cc77c21..7e348ab53 100644 --- a/public/tags/notes/page/2/index.html +++ b/public/tags/notes/page/2/index.html @@ -36,7 +36,7 @@ - + @@ -69,6 +69,14 @@ Home + CGSpace Notes + + Notes + + Posts + + Tags + @@ -206,7 +214,7 @@