Fetch website from its own git repository

Reviewed-by: Aleix Boné <abonerib@bsc.es>

doc/trim.sh

@@ -0,0 +1,46 @@
+#!/bin/sh
+
+# Trims the jungle repository by moving the website to its own repository and
+# removing it from jungle. It also removes big pdf files and kernel
+# configurations so the jungle repository is small.
+
+set -e
+
+if [ -e oldjungle -o -e newjungle -o -e website ]; then
+  echo "remove oldjungle/, newjungle/ and website/ first"
+  exit 1
+fi
+
+# Clone the old jungle repo
+git clone gitea@tent:rarias/jungle.git oldjungle
+
+# First split the website into a new repository
+mkdir website && git -C website init -b master
+git-filter-repo \
+  --path web \
+  --subdirectory-filter web \
+  --source oldjungle \
+  --target website
+
+# Then remove the website, pdf files and big kernel configs
+mkdir newjungle && git -C newjungle init -b master
+git-filter-repo \
+  --invert-paths \
+  --path web \
+  --path-glob 'doc*.pdf' \
+  --path-glob '**/kernel/configs/lockdep' \
+  --path-glob '**/kernel/configs/defconfig' \
+  --source oldjungle \
+  --target newjungle
+
+set -x
+
+du -sh oldjungle newjungle website
+#  57M  oldjungle
+# 2,3M  newjungle
+# 6,4M  website
+
+du -sh --exclude=.git oldjungle newjungle website
+#  30M  oldjungle
+# 700K  newjungle
+# 3,5M  website

m/apex/configuration.nix

@@ -5,6 +5,7 @@
     ../common/xeon.nix
     ../common/ssf/hosts.nix
     ../module/ceph.nix
+    ../module/hut-substituter.nix
     ../module/slurm-server.nix
     ./nfs.nix
     ./wireguard.nix
@@ -65,10 +66,4 @@
       iptables -I nixos-fw 2 -p tcp -s 84.88.52.176 -j nixos-fw-refuse
     '';
   };
-
-  # Use tent for cache
-  nix.settings = {
-    extra-substituters = [ "https://jungle.bsc.es/cache" ];
-    extra-trusted-public-keys = [ "jungle.bsc.es:pEc7MlAT0HEwLQYPtpkPLwRsGf80ZI26aj29zMw/HH0=" ];
-  };
 }

m/bay/configuration.nix

@@ -3,6 +3,7 @@
 {
   imports = [
     ../common/ssf.nix
+    ../module/hut-substituter.nix
     ../module/monitoring.nix
   ];
 

m/common/base/users.nix

@@ -156,18 +156,30 @@
       };
 
       csiringo = {
-        # Arbitrary UID but large so it doesn't collide with other users on ssfhead.
         uid = 9653;
         isNormalUser = true;
         home = "/home/Computational/csiringo";
         description = "Cesare Siringo";
         group = "Computational";
-        hosts = [ "apex" "weasel" ];
+        hosts = [ ];
         hashedPassword = "$6$0IsZlju8jFukLlAw$VKm0FUXbS.mVmPm3rcJeizTNU4IM5Nmmy21BvzFL.cQwvlGwFI1YWRQm6gsbd4nbg47mPDvYkr/ar0SlgF6GO1";
         openssh.authorizedKeys.keys = [
           "ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIHA65zvvG50iuFEMf+guRwZB65jlGXfGLF4HO+THFaed csiringo@bsc.es"
         ];
       };
+
+      acinca = {
+        uid = 9654;
+        isNormalUser = true;
+        home = "/home/Computational/acinca";
+        description = "Arnau Cinca";
+        group = "Computational";
+        hosts = [ "apex" "hut" "fox" "owl1" "owl2" ];
+        hashedPassword = "$6$S6PUeRpdzYlidxzI$szyvWejQ4hEN76yBYhp1diVO5ew1FFg.cz4lKiXt2Idy4XdpifwrFTCIzLTs5dvYlR62m7ekA5MrhcVxR5F/q/";
+        openssh.authorizedKeys.keys = [
+          "ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIFmMqKqPg4uocNOr3O41kLbZMOMJn3m2ZdN1JvTR96z3 bsccns@arnau-bsc"
+        ];
+      };
     };
 
     groups = {

m/eudy/configuration.nix

@@ -9,6 +9,7 @@
     ./cpufreq.nix
     ./fs.nix
     ./users.nix
+    ../module/hut-substituter.nix
     ../module/debuginfod.nix
   ];
 

m/fox/configuration.nix

@@ -8,6 +8,7 @@
     ../module/emulation.nix
     ../module/nvidia.nix
     ../module/slurm-client.nix
+    ../module/hut-substituter.nix
     ./wireguard.nix
   ];
 
@@ -62,12 +63,6 @@
     interfaces.enp1s0f0np0.useDHCP = true;
   };
 
-  # Use hut for cache
-  nix.settings = {
-    extra-substituters = [ "https://jungle.bsc.es/cache" ];
-    extra-trusted-public-keys = [ "jungle.bsc.es:pEc7MlAT0HEwLQYPtpkPLwRsGf80ZI26aj29zMw/HH0=" ];
-  };
-
   # Recommended for new graphics cards
   hardware.nvidia.open = true;
 

m/hut/nginx.nix

@@ -2,10 +2,13 @@
 let
   website = pkgs.stdenv.mkDerivation {
     name = "jungle-web";
-    src = theFlake;
+    src = pkgs.fetchgit {
+      url = "https://jungle.bsc.es/git/rarias/jungle-website.git";
+      rev = "739bf0175a7f05380fe7ad7023ff1d60db1710e1";
+      hash = "sha256-ea5DzhYTzZ9TmqD+x95rdNdLbxPnBluqlYH2NmBYmc4=";
+    };
     buildInputs = [ pkgs.hugo ];
     buildPhase = ''
-      cd web
       rm -rf public/
       hugo
     '';

m/lake2/configuration.nix

@@ -4,6 +4,7 @@
   imports = [
     ../common/ssf.nix
     ../module/monitoring.nix
+    ../module/hut-substituter.nix
   ];
 
   boot.loader.grub.device = "/dev/disk/by-id/wwn-0x55cd2e414d53563a";

m/module/hut-substituter.nix

@@ -6,5 +6,8 @@
     {
       extra-substituters = [ "http://hut/cache" ];
       extra-trusted-public-keys = [ "jungle.bsc.es:pEc7MlAT0HEwLQYPtpkPLwRsGf80ZI26aj29zMw/HH0=" ];
+
+      # Set a low timeout in case hut is down
+      connect-timeout = 3; # seconds
     };
 }

m/module/slurm-client.nix

@@ -12,6 +12,12 @@
     # https://github.com/NixOS/nixpkgs/commit/ae93ed0f0d4e7be0a286d1fca86446318c0c6ffb
     # https://bugs.schedmd.com/show_bug.cgi?id=2095#c24
     KillMode = lib.mkForce "control-group";
+
+    # If slurmd fails to contact the control server it will fail, causing the
+    # node to remain out of service until manually restarted. Always try to
+    # restart it.
+    Restart = "always";
+    RestartSec = "30s";
   };
 
   services.slurm.client.enable = true;

m/raccoon/configuration.nix

@@ -9,6 +9,7 @@
     ../module/nvidia.nix
     ../eudy/kernel/perf.nix
     ./wireguard.nix
+    ../module/hut-substituter.nix
   ];
 
   # Don't install Grub on the disk yet
@@ -51,11 +52,6 @@
     options = [ "nfsvers=3" "rsize=1024" "wsize=1024" "cto" "nofail" ];
   };
 
-  nix.settings = {
-    extra-substituters = [ "https://jungle.bsc.es/cache" ];
-    extra-trusted-public-keys = [ "jungle.bsc.es:pEc7MlAT0HEwLQYPtpkPLwRsGf80ZI26aj29zMw/HH0=" ];
-  };
-
   # Enable performance governor
   powerManagement.cpuFreqGovernor = "performance";
 

m/tent/configuration.nix

@@ -15,6 +15,7 @@
     ../hut/msmtp.nix
     ../module/p.nix
     ../module/vpn-dac.nix
+    ../module/hut-substituter.nix
   ];
 
   # Select the this using the ID to avoid mismatches

m/tent/nginx.nix

@@ -2,10 +2,13 @@
 let
   website = pkgs.stdenv.mkDerivation {
     name = "jungle-web";
-    src = theFlake;
+    src = pkgs.fetchgit {
+      url = "https://jungle.bsc.es/git/rarias/jungle-website.git";
+      rev = "739bf0175a7f05380fe7ad7023ff1d60db1710e1";
+      hash = "sha256-ea5DzhYTzZ9TmqD+x95rdNdLbxPnBluqlYH2NmBYmc4=";
+    };
     buildInputs = [ pkgs.hugo ];
     buildPhase = ''
-      cd web
       rm -rf public/
       hugo
     '';

m/weasel/configuration.nix

@@ -3,6 +3,7 @@
 {
   imports = [
     ../common/ssf.nix
+    ../module/hut-substituter.nix
   ];
 
   # Select this using the ID to avoid mismatches

web/content/posts/2025-09-26/_index.md

@@ -0,0 +1,49 @@
+---
+title: "Update 2025-09-26"
+author: "Rodrigo Arias Mallo"
+date: 2025-09-26
+---
+
+This is a summary of notable changes introduced in the last two years. We
+continue to maintain all machines updated to the last NixOS release (currently
+NixOS 25.05).
+
+### New compute node: fox
+
+We have a new [fox machine](/fox), with two AMD Genoa 9684X CPUs and two NVIDIA
+RTX4000 GPUs. During the last months we have been doing some tests and it seems
+that most of the components work well. We have configured CUDA to use the NVIDIA
+GPUs, as well as AMD uProf to trace performance and energy counters from the
+CPUs.
+
+### Upgraded login node: apex
+
+We have upgraded the operating system on the login node to NixOS, which now runs
+Linux 6.15.6. During the upgrade, we have detected a problem with the storage
+disks. The `/` and `/home` partitions sit on a
+[RAID 5](https://en.wikipedia.org/wiki/Standard_RAID_levels#RAID_5),
+transparently handled by a RAID hardware controller which starts its own
+firmware before passing the control to the BIOS to continue the boot sequence. A
+problem during the startup of the firmware prevented the node to even reach the
+BIOS screen.
+
+After a long debugging session, we detected that the flash memory that stores
+the firmware of the hardware controller was likely to be the issue, since
+[memory cells](https://en.wikipedia.org/wiki/Flash_memory#Principles_of_operation)
+may lose charge over time and can end up corrupting the content. We flashed
+the latest firmware so the memory cells are charged again with the new bits and
+that fixed the problem. Hopefully we will be able to use it for some more years.
+
+The SLURM server has been moved to apex which allows users to also submit jobs
+to fox.
+
+### Migrated machines to BSC building
+
+The server room had a temperature issue that had been affecting our machines
+since the end of February of 2025. As the summer approached, the temperature
+exceeded the safe limits for our hardware, so we had to shutdown the cluster.
+
+![Room temperature](temp.png)
+
+Since then, we have moved the cluster to BSC premises, where it now rests at a
+stable temperature, so hopefully we won't have more unscheduled downtime.