diff --git a/content/posts/2019-03.md b/content/posts/2019-03.md
index c8dc557a5..418543e6a 100644
--- a/content/posts/2019-03.md
+++ b/content/posts/2019-03.md
@@ -734,5 +734,70 @@ $ grep 'Can not load requested doc' cocoon.log.2019-03-25 | grep -oE '2019-03-25
- By default you get the Commons DBCP one unless you specify factory `org.apache.tomcat.jdbc.pool.DataSourceFactory`
- Now I see all my interceptor settings etc in jconsole, where I didn't see them before (also a new `tomcat.jdbc` mbean)!
- No wonder our settings didn't quite match the ones in the [Tomcat DBCP Pool docs](https://tomcat.apache.org/tomcat-7.0-doc/jdbc-pool.html)
+- Uptime Robot reported that CGSpace went down and I see the load is very high
+- The top IPs around the time in the nginx API and web logs were:
+
+```
+# zcat --force /var/log/nginx/{oai,rest,statistics}.log /var/log/nginx/{oai,rest,statistics}.log.1 | grep -E "25/Mar/2019:(18|19|20|21)" | awk '{print $1}' | sort | uniq -c | sort -n | tail -n 10
+ 9 190.252.43.162
+ 12 157.55.39.140
+ 18 157.55.39.54
+ 21 66.249.66.211
+ 27 40.77.167.185
+ 29 138.220.87.165
+ 30 157.55.39.168
+ 36 157.55.39.9
+ 50 52.23.239.229
+ 2380 45.5.186.2
+# zcat --force /var/log/nginx/{access,error,library-access}.log /var/log/nginx/{access,error,library-access}.log.1 | grep -E "25/Mar/2019:(18|19|20|21)" | awk '{print $1}' | sort | uniq -c | sort -n | tail -n 10
+ 354 18.195.78.144
+ 363 190.216.179.100
+ 386 40.77.167.185
+ 484 157.55.39.168
+ 507 157.55.39.9
+ 536 2a01:4f8:140:3192::2
+ 1123 66.249.66.211
+ 1186 93.179.69.74
+ 1222 35.174.184.209
+ 1720 2a01:4f8:13b:1296::2
+```
+- The IPs look pretty normal except we've never seen `93.179.69.74` before, and it uses the following user agent:
+
+```
+Mozilla/5.0 (Windows NT 5.1) AppleWebKit/535.1 (KHTML, like Gecko) Chrome/13.0.782.20 Safari/535.1
+```
+
+- Surprisingly they are re-using their Tomcat session:
+
+```
+$ grep -o -E 'session_id=[A-Z0-9]{32}:ip_addr=93.179.69.74' dspace.log.2019-03-25 | sort | uniq | wc -l
+1
+```
+
+- That's weird because the total number of sessions today seems low compared to recent days:
+
+```
+$ grep -o -E 'session_id=[A-Z0-9]{32}' dspace.log.2019-03-25 | sort -u | wc -l
+5657
+$ grep -o -E 'session_id=[A-Z0-9]{32}' dspace.log.2019-03-24 | sort -u | wc -l
+17710
+$ grep -o -E 'session_id=[A-Z0-9]{32}' dspace.log.2019-03-23 | sort -u | wc -l
+17179
+$ grep -o -E 'session_id=[A-Z0-9]{32}' dspace.log.2019-03-22 | sort -u | wc -l
+7904
+```
+
+- PostgreSQL seems to be pretty busy:
+
+```
+$ psql -c 'select * from pg_stat_activity' | grep -o -E '(dspaceWeb|dspaceApi|dspaceCli)' | sort | uniq -c
+ 11 dspaceApi
+ 10 dspaceCli
+ 67 dspaceWeb
+```
+
+- I restarted Tomcat and deployed the new Tomcat JDBC settings on CGSpace since I had to restart the server anyways
+ - I need to watch this carefully though because I've read some places that Tomcat's DBCP doesn't track statements and might create memory leaks if an application doesn't close statements before a connection gets returned back to the pool
+- According the Uptime Robot the server was up and down a few more times over the next hour so I restarted Tomcat again
diff --git a/docs/2019-03/index.html b/docs/2019-03/index.html
index 0becf1623..039929eff 100644
--- a/docs/2019-03/index.html
+++ b/docs/2019-03/index.html
@@ -25,7 +25,7 @@ I think I will need to ask Udana to re-copy and paste the abstracts with more ca
-
+
@@ -55,9 +55,9 @@ I think I will need to ask Udana to re-copy and paste the abstracts with more ca
"@type": "BlogPosting",
"headline": "March, 2019",
"url": "https://alanorth.github.io/cgspace-notes/2019-03/",
- "wordCount": "4722",
+ "wordCount": "5067",
"datePublished": "2019-03-01T12:16:30+01:00",
- "dateModified": "2019-03-25T12:39:22+02:00",
+ "dateModified": "2019-03-25T12:59:24+02:00",
"author": {
"@type": "Person",
"name": "Alan Orth"
@@ -1000,6 +1000,80 @@ org.postgresql.util.PSQLException: This statement has been closed.
Now I see all my interceptor settings etc in jconsole, where I didn’t see them before (also a new tomcat.jdbc mbean)!
No wonder our settings didn’t quite match the ones in the Tomcat DBCP Pool docs
+Uptime Robot reported that CGSpace went down and I see the load is very high
+The top IPs around the time in the nginx API and web logs were:
+
+
+# zcat --force /var/log/nginx/{oai,rest,statistics}.log /var/log/nginx/{oai,rest,statistics}.log.1 | grep -E "25/Mar/2019:(18|19|20|21)" | awk '{print $1}' | sort | uniq -c | sort -n | tail -n 10
+ 9 190.252.43.162
+ 12 157.55.39.140
+ 18 157.55.39.54
+ 21 66.249.66.211
+ 27 40.77.167.185
+ 29 138.220.87.165
+ 30 157.55.39.168
+ 36 157.55.39.9
+ 50 52.23.239.229
+ 2380 45.5.186.2
+# zcat --force /var/log/nginx/{access,error,library-access}.log /var/log/nginx/{access,error,library-access}.log.1 | grep -E "25/Mar/2019:(18|19|20|21)" | awk '{print $1}' | sort | uniq -c | sort -n | tail -n 10
+ 354 18.195.78.144
+ 363 190.216.179.100
+ 386 40.77.167.185
+ 484 157.55.39.168
+ 507 157.55.39.9
+ 536 2a01:4f8:140:3192::2
+ 1123 66.249.66.211
+ 1186 93.179.69.74
+ 1222 35.174.184.209
+ 1720 2a01:4f8:13b:1296::2
+
+
+
+- The IPs look pretty normal except we’ve never seen
93.179.69.74 before, and it uses the following user agent:
+
+
+Mozilla/5.0 (Windows NT 5.1) AppleWebKit/535.1 (KHTML, like Gecko) Chrome/13.0.782.20 Safari/535.1
+
+
+
+- Surprisingly they are re-using their Tomcat session:
+
+
+$ grep -o -E 'session_id=[A-Z0-9]{32}:ip_addr=93.179.69.74' dspace.log.2019-03-25 | sort | uniq | wc -l
+1
+
+
+
+- That’s weird because the total number of sessions today seems low compared to recent days:
+
+
+$ grep -o -E 'session_id=[A-Z0-9]{32}' dspace.log.2019-03-25 | sort -u | wc -l
+5657
+$ grep -o -E 'session_id=[A-Z0-9]{32}' dspace.log.2019-03-24 | sort -u | wc -l
+17710
+$ grep -o -E 'session_id=[A-Z0-9]{32}' dspace.log.2019-03-23 | sort -u | wc -l
+17179
+$ grep -o -E 'session_id=[A-Z0-9]{32}' dspace.log.2019-03-22 | sort -u | wc -l
+7904
+
+
+
+- PostgreSQL seems to be pretty busy:
+
+
+$ psql -c 'select * from pg_stat_activity' | grep -o -E '(dspaceWeb|dspaceApi|dspaceCli)' | sort | uniq -c
+ 11 dspaceApi
+ 10 dspaceCli
+ 67 dspaceWeb
+
+
+
+- I restarted Tomcat and deployed the new Tomcat JDBC settings on CGSpace since I had to restart the server anyways
+
+
+- I need to watch this carefully though because I’ve read some places that Tomcat’s DBCP doesn’t track statements and might create memory leaks if an application doesn’t close statements before a connection gets returned back to the pool
+
+- According the Uptime Robot the server was up and down a few more times over the next hour so I restarted Tomcat again
diff --git a/docs/sitemap.xml b/docs/sitemap.xml
index 4cee531ca..a45eb37d0 100644
--- a/docs/sitemap.xml
+++ b/docs/sitemap.xml
@@ -4,7 +4,7 @@
https://alanorth.github.io/cgspace-notes/2019-03/
- 2019-03-25T12:39:22+02:00
+ 2019-03-25T12:59:24+02:00
@@ -214,7 +214,7 @@
https://alanorth.github.io/cgspace-notes/
- 2019-03-25T12:39:22+02:00
+ 2019-03-25T12:59:24+02:00
0
@@ -225,7 +225,7 @@
https://alanorth.github.io/cgspace-notes/tags/notes/
- 2019-03-25T12:39:22+02:00
+ 2019-03-25T12:59:24+02:00
0
@@ -237,13 +237,13 @@
https://alanorth.github.io/cgspace-notes/posts/
- 2019-03-25T12:39:22+02:00
+ 2019-03-25T12:59:24+02:00
0
https://alanorth.github.io/cgspace-notes/tags/
- 2019-03-25T12:39:22+02:00
+ 2019-03-25T12:59:24+02:00
0