Merge branch 'master' of https://github.com/filguard/lotus-ops

minerdao · Aug 12, 2021 · 837e2b3 · 837e2b3
2 parents 612999c + 7b9ee58
commit 837e2b3
Show file tree

Hide file tree

Showing 31 changed files with 3,899 additions and 72 deletions.
diff --git a/README.md b/README.md
@@ -20,8 +20,5 @@
 - [Lotus日常运维CheckList](./documents/lotus-ops-checklist.md)
 - [常用环境变量说明](./documents/environment-usage.md)
 - [自动Pledge脚本使用说明](./documents/auto-pledge.md)
-- 时空证明运行流程、原理详解
-- 爆块逻辑及优化方法
-- 消息池清理操作
-- 源码编译及编译常见问题
 - 常见运维问题定位与解决
+  - [顽固扇区如何删除？](./documents/questions.md#1-顽固扇区如何删除)
diff --git a/ansible/2-start-window-post-miner.yaml → ansible/1-start-window-post-miner.yaml b/ansible/2-start-window-post-miner.yaml → ansible/1-start-window-post-miner.yaml
@@ -1,14 +1,13 @@
 - hosts: window-post-miner
   remote_user: '{{ remote_user }}'
   environment:
-    RUST_LOG: '{{ rust_log }}'
-    RUST_BACKTRACE: '{{ rust_backtrace }}'
     FIL_PROOFS_PARAMETER_CACHE: '{{ fil_proofs_parameter_cache }}'
     FIL_PROOFS_MAXIMIZE_CACHING: '{{ fil_proofs_maximize_caching }}'
     FIL_PROOFS_USE_GPU_COLUMN_BUILDER: '{{ fil_proofs_use_gpu_column_builder }}'
     FIL_PROOFS_USE_GPU_TREE_BUILDER: '{{ fil_proofs_use_gpu_tree_builder }}'
     FULLNODE_API_INFO: '{{ fullnode_api_info }}'
     LOTUS_PATH: '{{ lotus_path }}'
+    TMPDIR: '{{ tmp_dir }}'
     LOTUS_MINER_PATH: '{{ lotus_miner_path }}'
     TMPDIR: '{{ tmp_dir }}'
   tasks:

diff --git a/ansible/1-start-winning-post-miner.yaml → ansible/2-start-winning-post-miner.yaml b/ansible/1-start-winning-post-miner.yaml → ansible/2-start-winning-post-miner.yaml
@@ -9,6 +9,7 @@
     FIL_PROOFS_USE_GPU_TREE_BUILDER: '{{ fil_proofs_use_gpu_tree_builder }}'
     FULLNODE_API_INFO: '{{ fullnode_api_info }}'
     LOTUS_PATH: '{{ lotus_path }}'
+    TMPDIR: '{{ tmp_dir }}'
     LOTUS_MINER_PATH: '{{ lotus_miner_path }}'
     TMPDIR: '{{ tmp_dir }}'
   tasks:

diff --git a/ansible/5-start-seal-miner.yaml b/ansible/5-start-seal-miner.yaml
@@ -9,6 +9,7 @@
     FIL_PROOFS_USE_GPU_TREE_BUILDER: '{{ fil_proofs_use_gpu_tree_builder }}'
     FULLNODE_API_INFO: '{{ fullnode_api_info }}'
     LOTUS_PATH: '{{ lotus_path }}'
+    TMPDIR: '{{ tmp_dir }}'
     LOTUS_MINER_PATH: '{{ lotus_miner_path }}'
     TMPDIR: '{{ tmp_dir }}'
   tasks:

diff --git a/ansible/6-start-precommit-worker.yaml b/ansible/6-start-precommit-worker.yaml
@@ -1,9 +1,10 @@
-- hosts: precommit-worker
+- hosts: 10.0.1.16
   remote_user: '{{ remote_user }}'
   environment:
     RUST_LOG: '{{ rust_log }}'
     RUST_BACKTRACE: '{{ rust_backtrace }}'
     FIL_PROOFS_PARAMETER_CACHE: '{{ fil_proofs_parameter_cache }}'
+    FIL_PROOFS_PARENT_CACHE: '{{ fil_proofs_parent_cache }}'
     FIL_PROOFS_MAXIMIZE_CACHING: '{{ fil_proofs_maximize_caching }}'
     FIL_PROOFS_USE_GPU_COLUMN_BUILDER: '{{ fil_proofs_use_gpu_column_builder }}'
     FIL_PROOFS_USE_GPU_TREE_BUILDER: '{{ fil_proofs_use_gpu_tree_builder }}'
@@ -14,7 +15,7 @@
     - debug: var=ansible_all_ipv4_addresses[0]
     - debug: var=ansible_hostname
     - debug: var=lotus_miner_path
-
+    
     - name: Create cache path, lotus worker path, tmp path, log path
       file:
         path: '{{ item }}'
@@ -24,25 +25,43 @@
         mode: u=rwxr,g=xr,o=x
       loop:
         - '{{ fil_proofs_parent_cache }}'
-        - '{{ lotus_worker_path }}'
-        - '{{ tmp_dir }}/tmp'
+        - '{{ lotus_worker_path_0 }}'
+        - '{{ lotus_worker_path_1 }}'
+        - '{{ tmp_dir }}'
         - '{{ log_path }}'
 
     - name: Copy myscheduler config to precommit worker
       copy:
         src: '{{ workspace }}/lotus-ops/config/myscheduler-precommit.json'
         dest: '{{ item }}/myscheduler.json'
       loop:
-        - '{{ lotus_worker_path }}'
+        - '{{ lotus_worker_path_0 }}'
+        - '{{ lotus_worker_path_1 }}'
 
     - name: Start precommit worker 0
       shell: |
-        tmux new -s lotus -d -n worker
-        tmux send-keys -t lotus:worker "export LOTUS_WORKER_PATH={{ lotus_worker_path }}" C-m
-        tmux send-keys -t lotus:worker "export CUDA_VISIBLE_DEVICES=0" C-m
-        tmux send-keys -t lotus:worker "export TMPDIR={{ tmp_dir }}/tmp" C-m
-        tmux send-keys -t lotus:worker "export FIL_PROOFS_MULTICORE_SDR_PRODUCERS=1" C-m
-        tmux send-keys -t lotus:worker "export ENV_CPU_CORE_BEGIN_NUM=0" C-m
-        tmux send-keys -t lotus:worker "export ENV_CPU_CORE_END_NUM=16" C-m
-        tmux send-keys -t lotus:worker "lotus-worker run --listen={{ ansible_all_ipv4_addresses[0] }}:{{ worker_port }} --commit=false" C-m
-        tmux pipe-pane -o "cat >>{{ log_path }}/lotus-worker-`date +%Y-%m-%d-%H-%M`.log"
+        tmux new -s lotus -d -n worker0
+        tmux send-keys -t lotus:worker0 "export LOTUS_WORKER_PATH={{ lotus_worker_path_0 }}" C-m
+        tmux send-keys -t lotus:worker0 "export CUDA_VISIBLE_DEVICES=0" C-m
+        tmux send-keys -t lotus:worker0 "export TMPDIR={{ tmp_dir }}" C-m
+        tmux send-keys -t lotus:worker0 "export FIL_PROOFS_MULTICORE_SDR_PRODUCERS=1" C-m
+        tmux send-keys -t lotus:worker0 "export ENV_CPU_CORE_BEGIN_NUM=0" C-m
+        tmux send-keys -t lotus:worker0 "export ENV_CPU_CORE_END_NUM=8" C-m
+        tmux send-keys -t lotus:worker0 "echo $ENV_CPU_CORE_BEGIN_NUM" C-m
+        tmux send-keys -t lotus:worker0 "echo $ENV_CPU_CORE_END_NUM" C-m
+        tmux send-keys -t lotus:worker0 "lotus-worker run --listen={{ ansible_all_ipv4_addresses[0] }}:{{ lotus_worker_port_0 }} --commit=false" C-m
+        tmux pipe-pane -o "cat >>{{ log_path }}/lotus-worker0-`date +%Y-%m-%d-%H-%M`.log"
+
+    - name: Start commit worker 1
+      shell: |
+        tmux new-window -t lotus -n worker1
+        tmux send-keys -t lotus:worker1 "export LOTUS_WORKER_PATH={{ lotus_worker_path_1 }}" C-m
+        tmux send-keys -t lotus:worker1 "export CUDA_VISIBLE_DEVICES=0" C-m
+        tmux send-keys -t lotus:worker1 "export TMPDIR={{ tmp_dir }}" C-m
+        tmux send-keys -t lotus:worker1 "export FIL_PROOFS_MULTICORE_SDR_PRODUCERS=1" C-m
+        tmux send-keys -t lotus:worker1 "export ENV_CPU_CORE_BEGIN_NUM=8" C-m
+        tmux send-keys -t lotus:worker1 "export ENV_CPU_CORE_END_NUM=16" C-m
+        tmux send-keys -t lotus:worker1 "echo $ENV_CPU_CORE_BEGIN_NUM" C-m
+        tmux send-keys -t lotus:worker1 "echo $ENV_CPU_CORE_END_NUM" C-m
+        tmux send-keys -t lotus:worker1 "lotus-worker run --listen={{ ansible_default_ipv4.address }}:{{ lotus_worker_port_1 }} --commit=false" C-m
+        tmux pipe-pane -o "cat >>{{ log_path }}/lotus-worker1-`date +%Y-%m-%d-%H-%M`.log"
diff --git a/ansible/7-start-commit-worker.yaml b/ansible/7-start-commit-worker.yaml
@@ -1,4 +1,4 @@
-- hosts: commit-worker
+- hosts: 10.0.2.13
   remote_user: '{{ remote_user }}'
   environment:
     RUST_LOG: '{{ rust_log }}'
@@ -54,4 +54,4 @@
         tmux send-keys -t lotus:worker1 "export CUDA_VISIBLE_DEVICES=1" C-m
         tmux send-keys -t lotus:worker1 "export TMPDIR={{ tmp_dir }}/tmp1" C-m
         tmux send-keys -t lotus:worker1 "lotus-worker run --listen={{ ansible_all_ipv4_addresses[0] }}:{{ lotus_worker_port_1 }} --addpiece=false --precommit1=false --precommit2=false" C-m
-        tmux pipe-pane -o "cat >>{{ log_path }}/lotus-worker1-`date +%Y-%m-%d-%H-%M`.log"
+        tmux pipe-pane -o "cat >>{{ log_path }}/lotus-worker1-`date +%Y-%m-%d-%H-%M`.log"
diff --git a/ansible/copy-bin-to-amd.yaml b/ansible/copy-bin-to-amd.yaml
@@ -5,7 +5,7 @@
   become_method: sudo
   vars:
     - lotus_bin_path: 'lotus-bin'
-    - version: '1.7.0'
+    - version: 'v1.10.0'
     - platform: 'amd'
   tasks:
     - name: Copy lotus bin to all amd
@@ -14,8 +14,8 @@
         dest: '/usr/local/bin/'
         mode: +x
       loop:
-        - '{{ workspace }}/{{ lotus_bin_path }}/{{ version }}/{{ platform }}/lotus'
-        - '{{ workspace }}/{{ lotus_bin_path }}/{{ version }}/{{ platform }}/lotus-miner'
-        - '{{ workspace }}/{{ lotus_bin_path }}/{{ version }}/{{ platform }}/lotus-worker'
-        - '{{ workspace }}/{{ lotus_bin_path }}/{{ version }}/{{ platform }}/lotus-shed'
-        - '{{ workspace }}/{{ lotus_bin_path }}/{{ version }}/{{ platform }}/lotus-bench'
+        - '{{ workspace }}/{{ lotus_bin_path }}/{{ version }}/lotus'
+        - '{{ workspace }}/{{ lotus_bin_path }}/{{ version }}/lotus-miner'
+        - '{{ workspace }}/{{ lotus_bin_path }}/{{ version }}/lotus-worker'
+        - '{{ workspace }}/{{ lotus_bin_path }}/{{ version }}/lotus-shed'
+        - '{{ workspace }}/{{ lotus_bin_path }}/{{ version }}/lotus-bench'
diff --git a/ansible/install-node-exporter.yaml b/ansible/install-node-exporter.yaml
@@ -1,4 +1,5 @@
-- hosts: public-daemon,winning-post-miner,window-post-miner,seal-miner,precommit-worker,commit-worker
+# - hosts: public-daemon,private-daemon,winning-post-miner,window-post-miner,seal-miner,precommit-worker,commit-worker
+- hosts: 10.0.1.17
   remote_user: '{{ remote_user }}'
   gather_facts: no
   become: yes
@@ -7,7 +8,7 @@
   tasks:
     - name: Create prometheus run direction
       file:
-        path: '/home/fil/disk_md0/prometheus/run'
+        path: '/home/fil/prometheus/run'
         state: directory
 
     - name: Copy node exporter script to remote server

diff --git a/ansible/lotus-version-check.yaml b/ansible/lotus-version-check.yaml
@@ -0,0 +1,28 @@
+- hosts: amd
+  tasks:
+    - debug: var=ansible_default_ipv4.address
+    - debug: var=ansible_hostname
+
+    - name: Create workspace directory
+      file:
+        path: '{{ workspace }}'
+        state: directory
+        owner: '{{ remote_user }}'
+        group: '{{ remote_user }}'
+        mode: +x
+
+    - name: Copy check script to remote server
+      copy:
+        src: '{{ workspace }}/lotus-ops/scripts/lotus-version-check.sh'
+        dest: '{{ workspace }}'
+        owner: '{{ remote_user }}'
+        group: '{{ remote_user }}'
+        mode: +x
+
+    - name: Lotus version check
+      shell: '{{ workspace }}/lotus-version-check.sh'
+      register: check_result
+
+    - name: Echo check result
+      debug:
+        msg: '{{ check_result.stdout }}'
diff --git a/ansible/worker-pre-check.yaml b/ansible/worker-pre-check.yaml
@@ -25,4 +25,4 @@
 
     - name: Echo check result
       debug:
-        msg: '{{ check_result.stdout }}'
+        msg: '{{ check_result.stdout }}'
diff --git a/config/Miner-monitoring.json b/config/Miner-monitoring.json
@@ -17,7 +17,7 @@
   "editable": true,
   "gnetId": 8919,
   "graphTooltip": 0,
-  "id": 1,
+  "id": null,
   "iteration": 1615014350433,
   "links": [
     {