EOS Graphical Monitoring

1

Dear All,
Is there any Graphical Tool for Monitoring of EOS?

Presently data from Kolkata::EOS and Kolkata::SE have been transferring to Kolkata::EOS2. In this respect we want to monitor the performance of transfers, disks, nodes etc under RAIN-6 configuration.

We are following via group, node, fs etc command.

[root@eos-mgm ~]# eos -b group ls --io
┌────────────────┬──────────┬────────────┬────────────┬──────────┬──────────┬──────────┬──────┬──────┬────────────┬────────────┬────────────┬───────────┬──────────┬──────────┐
│name │ diskload│ diskr-MB/s│ diskw-MB/s│ eth-MiB/s│ ethi-MiB│ etho-MiB│ ropen│ wopen│ used-bytes│ max-bytes│ used-files│ max-files│ bal-shd│ drain-shd│
└────────────────┴──────────┴────────────┴────────────┴──────────┴──────────┴──────────┴──────┴──────┴────────────┴────────────┴────────────┴───────────┴──────────┴──────────┘
default.0 0.00 0 18 8344 637 430 6 205 6.16 TB 68.56 TB 462.95 K 6.70 G 0 0
default.1 0.00 0 24 8344 637 430 2 232 4.80 TB 68.56 TB 376.80 K 6.70 G 0 0
default.10 0.07 86 41 8344 637 430 3 129 6.18 TB 68.56 TB 464.32 K 6.70 G 0 0
default.11 0.00 0 0 8344 637 430 2 174 4.81 TB 68.56 TB 377.24 K 6.70 G 0 0
default.12 0.00 0 6 8344 637 430 0 136 6.07 TB 68.56 TB 463.99 K 6.70 G 0 0
default.13 0.00 0 17 8344 637 430 6 235 4.81 TB 68.56 TB 376.69 K 6.70 G 0 0
default.14 0.00 0 10 8344 637 430 7 246 6.11 TB 68.56 TB 462.72 K 6.70 G 0 0
default.15 0.00 0 15 8344 637 430 0 201 4.73 TB 68.56 TB 377.77 K 6.70 G 0 0
default.2 0.00 0 11 8344 637 430 0 134 6.10 TB 68.56 TB 462.84 K 6.70 G 0 0
default.3 0.00 0 13 8344 637 430 1 168 4.92 TB 68.56 TB 377.00 K 6.70 G 0 0
default.4 0.02 47 20 8344 637 430 0 151 6.17 TB 68.56 TB 463.10 K 6.70 G 0 0
default.5 0.02 45 7 8344 637 430 0 157 4.88 TB 68.56 TB 378.38 K 6.70 G 0 0
default.6 0.04 91 11 8344 637 430 0 136 6.12 TB 68.56 TB 464.39 K 6.70 G 0 0
default.7 0.00 0 27 8344 637 430 3 203 5.08 TB 68.56 TB 376.76 K 6.70 G 0 0
default.8 0.01 0 73 8344 637 430 0 203 6.12 TB 68.56 TB 462.31 K 6.70 G 0 0
default.9 0.02 0 131 8344 637 430 2 159 4.93 TB 68.56 TB 378.13 K 6.70 G 0 0

Are all these information are written under logs also? if yes How to parse these info from logs.

Regards
Vikas

2

Hi Vikas,

You can give it a try a this prometheus exporter for eos that I’ve written some time ago:

https://gitlab.cern.ch/rvalverd/eos_exporter

Cheers,
Roberto

3

Dear Roberto,

We install your mentioned prometheus exporter in our EOS manager node (Headnode) at ALICE::KOLKATA::EOS2.
But it shows “panic:signal:killed” error as follows:-

================================
[root@eos-mgm eos_exporter]# ./eos_exporter eos-instance=“alicekolkata”
INFO[0000] Starting eos_exporter (version=, branch=, revision=) source=“eos_exporter.go:78”
INFO[0000] Build context (go=go1.12.1, user=, date=) source=“eos_exporter.go:79”
INFO[0000] Starting eos exporter for cluster: %s eos-instance=alicekolkata source=“eos_exporter.go:81”
INFO[0000] Listening on :9373 source=“eos_exporter.go:95”
panic: signal: killed

goroutine 22 [running]:
eos_exporter/collector.(*FSCollector).collectFSDF(0xc000001980, 0xb2b180, 0xc000010580)
/usr/local/go/src/eos_exporter/collector/fs.go:364 +0x1c28
eos_exporter/collector.(*FSCollector).Collect(0xc000001980, 0xc0001b2120)
/usr/local/go/src/eos_exporter/collector/fs.go:568 +0x2f
main.(*EOSExporter).Collect(0xc0001f66a0, 0xc0001b2120)
/usr/local/go/src/eos_exporter/eos_exporter.go:62 +0xa6
eos_exporter/vendor/github.com/prometheus/client_golang/prometheus.(*Registry).Gather.func2(0xc0001ae060, 0xc0001b2120, 0xb2c940, 0xc0001f66a0)
/usr/local/go/src/eos_exporter/vendor/github.com/prometheus/client_golang/prometheus/registry.go:433 +0x61
created by eos_exporter/vendor/github.com/prometheus/client_golang/prometheus.(*Registry).Gather
/usr/local/go/src/eos_exporter/vendor/github.com/prometheus/client_golang/prometheus/registry.go:431 +0x2e9
[root@eos-mgm eos_exporter]#

================================
We have install it in this following path:-
[root@eos-mgm eos_exporter]# pwd
/usr/local/go/src/eos_exporter

=================================
After running the code provided by you, EOS Exporter is running on the browser with port no 9373. But every time after some while it gets crashed and in terminal a message of “panic: signal: killed” appears.

Please let us know how can we rectify it?

(Note: OS in EOS is CentOS 7.6. Should we install it in FST also?)

Regards
Prasun

4

Hi prasun,

I will have a look and let you know,

Cheers,

EDIT: no, no need to install it in the fst’s.