Hello, I deleted a poller from the GUI (thrash can icon in poller menu)
How can I add it back?
Thank you very much for help
When it runs it will be added back (within 5 minutes).
1 Like
Hmm apparently the node is polling now but it does not show up in the GUI. I waited overnight.
Another question, i disabled polling on my main/gui-server. Validate php shows that server as:
[FAIL] Some poller nodes have not checked in recently
Inactive Nodes:
“main-server node”
That “fail” is probably false positive since that server is not supposed to do polling anyway?
Here are some poller-wrapper.logs from the poller not showing in GUI:
INFO :: worker Thread-7 finished device 543 in 51 seconds
INFO :: worker Thread-2 finished device 257 in 51 seconds
INFO :: worker Thread-4 finished device 544 in 51 seconds
INFO :: Registered as Master
INFO :: starting the poller check at 2025-06-27 10:00:02 with 32 threads
INFO :: The device 1551 appears to be being checked by another node
INFO :: The device 1460 appears to be being checked by another node
INFO :: The device 1228 appears to be being checked by another node
INFO :: The device 1225 appears to be being checked by another node
INFO :: 1 Node(s) Total
INFO :: 2 Node(s) Total
INFO :: worker Thread-44 finished device 2880 in 29 seconds
INFO :: worker Thread-46 finished device 2879 in 29 seconds
INFO :: worker Thread-45 finished device 3320 in 32 seconds
INFO :: worker Thread-18 finished device 1576 in 42 seconds
INFO :: worker Thread-15 finished device 1459 in 43 seconds
INFO :: worker Thread-14 finished device 1454 in 43 seconds
INFO :: worker Thread-12 finished device 1455 in 43 seconds
INFO :: worker Thread-9 finished device 983 in 45 seconds
INFO :: worker Thread-11 finished device 984 in 45 seconds
INFO :: worker Thread-8 finished device 961 in 45 seconds
INFO :: worker Thread-10 finished device 960 in 45 seconds
INFO :: worker Thread-7 finished device 542 in 51 seconds
INFO :: worker Thread-4 finished device 543 in 51 seconds
INFO :: worker Thread-6 finished device 545 in 51 seconds
INFO :: worker Thread-2 finished device 544 in 51 seconds
INFO :: worker Thread-3 finished device 257 in 51 seconds
INFO :: worker Thread-5 finished device 258 in 51 seconds
Validate PHP of the “lost node” which was deleted earlier
| Component | Version |
|---|---|
| LibreNMS | 25.6.0-63-g1bcd63f93 (2025-06-26T23:02:57+07:00) |
| DB Schema | 2025_06_11_173916_drop_alert_rule (344) |
| PHP | 8.3.22 |
| Python | 3.9.21 |
| Database | MySQL 8.0.41 |
| RRDTool | 1.7.2 |
| SNMP | 5.9.1 |
| =========================================== |
[OK] Composer Version: 2.8.9
[OK] Dependencies up-to-date.
[OK] Database Connected
[OK] Database Schema is current
[OK] SQL Server meets minimum requirements
[OK] lower_case_table_names is enabled
[OK] MySQL engine is optimal
[OK] Database and column collations are correct
[OK] Database schema correct
[OK] MySQL and PHP time match
[OK] Distributed Polling setting is enabled globally
[OK] Connection to memcached is ok
[OK] Connected to rrdcached
[OK] Active pollers found
[OK] Dispatcher Service not detected
[OK] Locks are functional
[FAIL] Some poller nodes have not checked in recently
Inactive Nodes:
main-server node <— Not supposed to poll, polling commented out on .py
[OK] Redis is unavailable
[OK] rrdtool version ok
[OK] Connected to rrdcached
Okay it’s not actually polling anything and not showing up in GUI. Yet the validate.php is fine on all servers.
I am at loss here. I don’t see any errors anywhere, schedulers are running but the node won’t join to cluster. I rebooted every server and my database and nothing.
Running the poller wrapper manually on the lost node it shows this:
./poller-wrapper.py 16
INFO :: This system is already joined as the service master.
Is this the problem? When i run the poller-wrapper manually on other nodes they point to the “lost node”
INFO :: Registered as Node joining Master b’lostnode.mydomain.com’
And when i manually run the wrapper.py on the “lost node” i get the error below:
Traceback (most recent call last):
File “/opt/librenms/./poller-wrapper.py”, line 70, in
wrapper.wrapper(
File “/opt/librenms/LibreNMS/wrapper.py”, line 390, in wrapper
if str(MEMC.get(MASTER_TAG)) == config[“distributed_poller_name”]:
File “/opt/librenms/.local/lib/python3.9/site-packages/memcache.py”, line 888, in get
return self._get(‘get’, key)
File “/opt/librenms/.local/lib/python3.9/site-packages/memcache.py”, line 872, in _get
return _unsafe_get()
File “/opt/librenms/.local/lib/python3.9/site-packages/memcache.py”, line 861, in _unsafe_get
value = self._recv_value(server, flags, rlen)
File “/opt/librenms/.local/lib/python3.9/site-packages/memcache.py”, line 1026, in _recv_value
return val
UnboundLocalError: local variable ‘val’ referenced before assignment
BUT when i execute flush on memcached:
echo “flush_all” | nc <memcached_server_ip> 11211
Then the wrapper runs (with few errors and polling every device) on the lost node and it appeared in the GUI.
I’ll keep digging…
There was a conflicting install of python3-memcached with DNF and PIP. Pip package uninstalled and the lost node is working now.
1 Like
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.