summaryrefslogtreecommitdiffstats
path: root/krebs/2configs/shack/prometheus/alert-rules.nix
diff options
context:
space:
mode:
authormakefu <github@syntax-fehler.de>2020-09-14 15:45:37 +0200
committermakefu <github@syntax-fehler.de>2020-09-14 15:45:37 +0200
commit88aee4087ebf00e32b39476ddbd82c7e69490280 (patch)
tree103e519f10a144677121c6555be554adad3607af /krebs/2configs/shack/prometheus/alert-rules.nix
parentdf4f3982b15bf107db1b2c5cad58167e30fb3dec (diff)
shack/prometheus: add urls to alerts
Diffstat (limited to 'krebs/2configs/shack/prometheus/alert-rules.nix')
-rw-r--r--krebs/2configs/shack/prometheus/alert-rules.nix6
1 files changed, 3 insertions, 3 deletions
diff --git a/krebs/2configs/shack/prometheus/alert-rules.nix b/krebs/2configs/shack/prometheus/alert-rules.nix
index 0da1128ab..1c2d0b1ad 100644
--- a/krebs/2configs/shack/prometheus/alert-rules.nix
+++ b/krebs/2configs/shack/prometheus/alert-rules.nix
@@ -13,7 +13,7 @@ in {
expr = ''(node_filesystem_avail_bytes{alias="wolf.shack",mountpoint="/"} * 100) / node_filesystem_size_bytes{alias="wolf.shack",mountpoint="/"} < ${disk_free_threshold}'';
labels.severity = "warning";
annotations.summary = "{{ $labels.alias }} root disk full";
- annotations.url = "http://grafana.shack/";
+ annotations.url = "http://grafana.shack/d/hb7fSE0Zz/shack-system-dashboard?orgId=1&var-job=node&var-hostname=All&var-node=wolf.shack:9100&var-device=All&var-maxmount=%2F&var-show_hostname=wolf";
annotations.description = ''The root disk of {{ $labels.alias }} has {{ $value | printf "%.2f" }}% free disk space (Threshold at ${disk_free_threshold}%).A vast number of shackspace services will stop working. CI for deploying new configuration will also seize working. Log in to the system and run `nix-collect-garbage -d` and clean up the shack share folder in `/home/share` .If this does not help you can check `du -hs /var/ | sort -h`, run `docker system prune` or if you are really desperate run `du -hs / | sort -h` and go through the folders recursively until you've found something to delete'';
}
{
@@ -22,7 +22,7 @@ in {
expr = ''(node_filesystem_avail_bytes{alias="puyak.shack",mountpoint="/"} * 100) / node_filesystem_size_bytes{alias="puyak.shack",mountpoint="/"} < ${disk_free_threshold}'';
labels.severity = "warning";
annotations.summary = "{{ $labels.alias }} root disk full";
- annotations.url = "http://grafana.shack/";
+ annotations.url = "http://grafana.shack/d/hb7fSE0Zz/shack-system-dashboard?orgId=1&var-job=node&var-hostname=All&var-node=wolf.shack:9100&var-device=All&var-maxmount=%2F&var-show_hostname=puyak";
annotations.description = ''The root disk of {{ $labels.alias }} has {{ $value | printf "%.2f" }}% free disk space (Threshold at ${disk_free_threshold}%).Prometheus will not be able to create new alerts and CI for deploying new configuration will also seize working. Log in to the system and run `nix-collect-garbage -d` and if this does not help you can check `du -hs /var/ | sort -h`, run `docker system prune` or if you are really desperate run `du -hs / | sort -h` and go through the folders recursively until you've found something to delete'';
}
{
@@ -31,7 +31,7 @@ in {
for = "5m";
labels.severity = "page";
annotations.summary = "Instance {{ $labels.alias }} down for 5 minutes";
- annotations.url = "http://grafana.shack/";
+ annotations.url = "http://grafana.shack/d/hb7fSE0Zz/shack-system-dashboard?orgId=1&var-job=node&var-hostname=All&var-node=wolf.shack:9100&var-device=All&var-maxmount=%2F&var-show_hostname=wolf";
annotations.description = ''Host {{ $labels.alias }} went down and has not been reconnected after 5 minutes. This is probably bad news, try to restart the host via naproxen ( http://naproxen.shack:8006 ). Wolf being down means that CI,glados automation, light management and a couple of other services will not work anymore.'';
}
];