ZBX red status (timed out)

[WebGreg]

Hello Guys.

I'm trying to install the first agent and I can't.

Server=192.168.1.5 (the agent on it is working fine) - Ubuntu 20
Host=192.168.1.128 - Windows 7

But getting red status:

Get value from agent failed: ZBX_TCP_READ() timed out


1. Lets check if port 10050 is open: netstat -an | findstr 10050

TCP 0.0.0.0:10050 0.0.0.0:0 LISTENING
TCP [::]:10050 [::]:0 LISTENING

2. Firewall and antivirus are disabled.

3. From server: telnet 192.168.1.128 10050

On the host side in agent log I get:

failed to accept an incoming connection: from 192.168.1.5: reading first byte from connection failed: [0x0000274C] A connection attempt failed because the connected party did not properly respond after a period of time, or established connection failed because connected host has failed to respond.

4. On the server: sudo netstat -ntlp | grep LISTEN

tcp 0 0 0.0.0.0:10050 0.0.0.0:* LISTEN 881/zabbix_agentd
tcp 0 0 0.0.0.0:10051 0.0.0.0:* LISTEN 1159/zabbix_server
tcp6 0 0 :::10050 :::* LISTEN 881/zabbix_agentd
tcp6 0 0 :::10051 :::* LISTEN

5. I have tried with the agent configuration in active and passive mode

6. The agent in active mode gave me: netstat -an | findstr 10050

TCP 0.0.0.0:10050 0.0.0.0:0 LISTENING
TCP 10.0.5.0:10050 192.168.1.5:56862 SYN_RECEIVED
TCP [::]:10050 [::]:0 LISTENING

But status is still red and no data recived.



Any hints?

[WebGreg]

The agent in active mode works, but still with red status.

By server side I tried with: nc -v -z 192.168.1.128 10050 command and:

Connection to 192.168.1.128 10050 port [tcp/zabbix-agent] succeeded!

[Hamardaban]

Read the documentation carefully - status is only relevant to passive checks
It's good that you checked the port availability and traffic flow. Next, see what the server and agent will write to the logs when trying to establish a connection .

[WebGreg]

Yes I know. And I want it in passive mode but works for me only active. Sorry for my english. What I meant was that if in active works, then communication is OK - so why isn't it passive?
I know there are differences in the direction of communication - what side initiating the connection and where what port needs to be opened. But I think the ports are open on both sides.

I tried disable active mode and try only in passive but no diffrence.

Log looks OK for me.


HOST:
6140:20210226:183508.912 using configuration file: C:\Program Files\Zabbix Agent\zabbix_agentd.conf
6140:20210226:183510.046 agent #0 started [main process]
2548:20210226:183510.048 agent #1 started [collector]
11780:20210226:183510.049 agent #2 started[listener #1]
11056:20210226:183510.050 agent #3 started[listener #2]
8872:20210226:183510.051 agent #4 started[listener #3]
12260:20210226:183510.052 agent #5 started[listener #4]
4580:20210226:183510.053 agent #6 started[listener #5]

before in active mode there was one line more:
10316:20210226:143657.954 agent #7 started [active checks #1]

SERVER:
822:20210226:143555.376 active check configuration update from [127.0.0.1:10051] is working again
823:20210226:143555.376 active check configuration update from [192.168.1.5:10051] is working again

[Hamardaban]

Please show the frontend picture with the agent settings and the agent config text ( remove the comments!)

[WebGreg]

HOST AGENT CONF:
LogFile=C:\Program Files\Zabbix Agent\zabbix_agentd.log
Server=192.168.1.5
ServerActive=192.168.1.5
Hostname=P2018-15-A
Include=C:\Program Files\Zabbix Agent\zabbix_agentd.conf.d\

SERVER CONF:
LogFile=/var/log/zabbix/zabbix_server.log
LogFileSize=100
PidFile=/run/zabbix/zabbix_server.pid
SocketDir=/run/zabbix
DBName=zabbix
DBUser=***********
DBPassword=*********
StartPollers=5
StartPollersUnreachable=5
CacheSize=128M
HistoryCacheSize=1024M
ValueCacheSize=128M
Timeout=10
StatsAllowedIP=127.0.0.1,192.168.1.128

SERVER AGENT CONF
LogFileSize=0
Server=127.0.0.1,192.168.1.5,192.168.1.128
ServerActive=127.0.0.1,192.168.1.5
Hostname=Zabbix_server

[Hamardaban]

in server.conf add SourceIP=192.168.1.5

in host agent.conf add (for example)
DenyKey=system.run[shutdown*]
AllowKey=system.run[*]

[WebGreg]

in server.conf add (...) in host agent.conf add

I added - no change.

Server=127.0.0.1,192.168.1.5,192.168.1.128 in server agent conf... that .128 address, It does not hurt, but it should not be there...

Yea... You know... whan I couldn't get a connection I started to mess I test every option that comes to my mind



Please tell me if i'm wrong...

1. ZBX is red because of passive mode.
2. In passive mode, the server initiates a connection on port 10050.
3. If I check from server side: nc -v -z host_ip 10050
and I will get an answer:
Connection to host_ip 10050 port [tcp/zabbix-agent] succeeded!
It means that nothing block connection and if ZBX is red it exclude the problem with network - it have to be something in config?

What does the status check? Some item? Maybe there is something wrong? I tired with "Windows by Zabbix agent active" and "Windows by Zabbix agent" templates.


Edit: No latest value data for Zabbix agent - agent.ping.

I tried with:

[B]# zabbix_get -k agent.ping -s TARGET_HOST[/B]

and "Command 'zabbix_get' not found"

Maybe this is the reason?

[B]# sudo apt install zabbix-get[/B]

Now zabbix_get returns 1 from server command line, but still without the latest data value.



And from host agent log (after "get value and test"):

3612:20210227:200629.910 Requested [agent.ping]
3612:20210227:200629.911 Sending back [1]
5212:20210227:200630.387 In collect_perfstat()
5212:20210227:200630.391 End of collect_perfstat()

if the agent receives the request and sends a response why I get timed out?

SOLVED - I changed TimeoutT=30 (before I had 10) in server.conf. But why does it take so long? Problem with server optimization?

Number of processes [avg]: 197.3348
Maximum number of processes [avg]: 4.19M
Trigger: Configured max number of processes is too low (<1024)

eth0 Bits recived [avg]: 123.78 Kbps
eth0 Bits sent [avg]: 378.83 Kbps

CPU Jumps:
Context switches per second [avg]: 3.77K
Interrupts per second [avg]: 669.9407
Load average [15m avg]: 1.2748
CPU utilization [avg]: 21.35 %
CPU iowait time [avg]: 13.63 % (max: 46.34%)

Free swap space [avg]: 3.83 GB

Disk average queue size [avg]: 0.5 (max 3.5)
Disk utilization [avg]: 45.93% (max: 99.36%)
Read request avg waiting time [avg]: 11.08ms
Write request avg waiting time [avg]: 3.5ms
Read rate [avg]: 3.24 r/s (max: 153.35)
Write rate [avg]: 249.75 w/s (1144.79)

Memory utilization [max] 24%

[cyber]

Server=127.0.0.1,192.168.1.5,192.168.1.128 in server agent conf... that .128 address, It does not hurt, but it should not be there...
And make up your mind, which interface you are going to use for servers agent config. Either localhost or real one... Confugure it in frontend accordingly and leave one IP in config. Having ListenIP parameter in server config helps to keep it also on one interface only.
Having multiple IP-s in config makes it check for config multiple times...