Servarica down?

tall_ice · March 25

Servarica down? Or extremely slow connection? or just me

tmagicturtle · March 25

Nope, my VPS with them is working just fine. Control panel/client area is all up for me, too

tall_ice · March 25

Hmm. I am definitely struggling to get in. I opened a ticket with them.

And once I get in. Command I type in are not responsive.

tenji · March 25

move your port to any random number. If only ssh were you using, it's ok to move the port.

tall_ice · March 25

Sorry. Why would moving port solve my connection issue? Not that I know for sure it's connection issue.

silicomnet · March 25

Network status https://status.servarica.com/
Submit a ticket, they reply fast.

tall_ice · March 25

@silicomnet said:
Network status https://status.servarica.com/
Submit a ticket, they reply fast.

Thanks for sharing this link. They did respond to my ticket saying they need to migrate me to another storage server that is less utilized.

Looks like it's just me!.

tenji · March 25

sorry if this not applied to you, but for me too many login bruteforcer on my ssh causing my ssh sessions so slow 😅
never experience overcrowded node, just noisy neighbour, with top command reveal steal above 10% enough to cause slow ssh response, ymmv.

silicomnet · March 25

@tenji said:
sorry if this not applied to you, but for me too many login bruteforcer on my ssh causing my ssh sessions so slow 😅
never experience overcrowded node, just noisy neighbour, with top command reveal steal above 10% enough to cause slow ssh response, ymmv.

run
netstat -tn 2>/dev/null | grep ":22 "

replace 22 with your ssh port and block range of ip, we have 2 backup servers with servarica and they respond very quickly.

terminator · March 25

Their storage servers share the cores. So maybe your node is overcrowded. You should once check your cpu utilization with w command

tall_ice · March 25

@silicomnet said:

@tenji said:
sorry if this not applied to you, but for me too many login bruteforcer on my ssh causing my ssh sessions so slow 😅
never experience overcrowded node, just noisy neighbour, with top command reveal steal above 10% enough to cause slow ssh response, ymmv.

run
netstat -tn 2>/dev/null | grep ":22 "

replace 22 with your ssh port and block range of ip, we have 2 backup servers with servarica and they respond very quickly.

They are responding but still in progress of migration, and connection is going up and down frequently.

If I am already using alternate port, do I still need to block ip range?

How can I tell how much steal there is?

Tasks: 116 total,   1 running, 115 sleeping,   0 stopped,   0 zombie
%Cpu(s):  0.5 us,  1.7 sy,  0.0 ni, 97.7 id,  0.0 wa,  0.0 hi,  0.0 si,  0.1 st 
MiB Mem :   2933.5 total,   1956.0 free,    519.9 used,    614.6 buff/cache     
MiB Swap:   1912.0 total,   1911.5 free,      0.5 used.   2413.6 avail Mem

tall_ice · March 25

@terminator said:
Their storage servers share the cores. So maybe your node is overcrowded. You should once check your cpu utilization with w command

That is what they are saying. This is the second time I have to move though.

kode9 · March 25

@tall_ice said:

@silicomnet said:
Network status https://status.servarica.com/
Submit a ticket, they reply fast.

Thanks for sharing this link. They did respond to my ticket saying they need to migrate me to another storage server that is less utilized.

Looks like it's just me!.

Ain't just you brother.
I have 2 services with them. The "polar storage" one is going up and down and up and down.
To these yabs results seem ok to you guys ?

## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ##

Yet-Another-Bench-Script

v2025-01-01

https://github.com/masonr/yet-another-bench-script

## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ##

Tue Mar 25 10:00:39 AM PDT 2025

Basic System Information:

Uptime : 0 days, 0 hours, 5 minutes
Processor : Intel(R) Xeon(R) Gold 6132 CPU @ 2.60GHz
CPU cores : 2 @ 2599.913 MHz
AES-NI : ✔ Enabled
VM-x/AMD-V : ❌ Disabled
RAM : 1.9 GiB
Swap : 0.0 KiB
Disk : 1.9 TiB
Distro : Ubuntu 22.04.5 LTS
Kernel : 5.15.0-134-generic
VM Type : XEN
IPv4/IPv6 : ✔ Online / ✔ Online

IPv6 Network Information:

ISP : Rica Web Services
ASN : AS26832 Rica Web Services
Host : Rica Web Services
Location : Montreal, Quebec (QC)
Country : Canada

fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/mapper/ubuntu--vg-lv--root):

Block Size	4k (IOPS)	64k (IOPS)
Read	4.36 MB/s (1.0k)	7.32 MB/s (114)
Write	4.38 MB/s (1.0k)	7.73 MB/s (120)
Total	8.74 MB/s (2.1k)	15.05 MB/s (234)

Block Size	512k (IOPS)	1m (IOPS)
------	--- ----	---- ----
Read	581.00 KB/s (1)	6.34 MB/s (6)
Write	632.00 KB/s (1)	6.45 MB/s (6)
Total	1.21 MB/s (2)	12.80 MB/s (12)

iperf3 Network Speed Tests (IPv4):

Provider | Location (Link) | Send Speed | Recv Speed | Ping
----- | ----- | ---- | ---- | ----
Clouvider | London, UK (10G) | 3.96 Mbits/sec | 1.44 Gbits/sec | 80.0 ms
Eranium | Amsterdam, NL (100G) | 4.26 Mbits/sec | 2.16 Gbits/sec | 83.2 ms
Uztelecom | Tashkent, UZ (10G) | 3.62 Mbits/sec | 826 Mbits/sec | 186 ms
Leaseweb | Singapore, SG (10G) | 2.11 Mbits/sec | 672 Mbits/sec | 244 ms
Clouvider | Los Angeles, CA, US (10G) | 5.58 Mbits/sec | 1.89 Gbits/sec | 73.8 ms
Leaseweb | NYC, NY, US (10G) | 31.7 Mbits/sec | 7.40 Gbits/sec | 9.24 ms
Performing IPv4 iperf3 recv test from Edgoo (Attempt #1 of 3)...

PS: that's all that I got `till it went down again.

jperkins · March 25

@tall_ice said:

@terminator said:
Their storage servers share the cores. So maybe your node is overcrowded. You should once check your cpu utilization with w command

That is what they are saying. This is the second time I have to move though.

Had servarica storage product ( the unlimited expanding storage ) for several years. Was moved around a couple times due to sluggishness and/or poor disk throughput. To be fair I didnt want one big disk of storage on / so my layout was not their standard, Their support was responsive and things would get better for awhile. Would still be with them but another player came on the storage scene and I moved on

tenji · March 25

@tall_ice said:

They are responding but still in progress of migration, and connection is going up and down frequently.

If I am already using alternate port, do I still need to block ip range?

no since they usually automated things, since I move it I rarely got hit with login attempt(s).

How can I tell how much steal there is?

Tasks: 116 total,   1 running, 115 sleeping,   0 stopped,   0 zombie
%Cpu(s):  0.5 us,  1.7 sy,  0.0 ni, 97.7 id,  0.0 wa,  0.0 hi,  0.0 si,  0.1 st 
MiB Mem :   2933.5 total,   1956.0 free,    519.9 used,    614.6 buff/cache     
MiB Swap:   1912.0 total,   1911.5 free,      0.5 used.   2413.6 avail Mem

the last indicator on %cpu(s), st = steal
everything on the left side of st were your internal vps (that I know, cmiiw), the st were indicator of resource used by hypervisor & other vps in the node.
try to open 2 login window / putty, one to watch the top, other one to check the responsiveness when you typing the command(s). depends on cpu type (+count) & node load, but on single vcpu vps 0.1 st were little annoying, but 4 vcpu 0.2-0.5 st were breeze, in my experience. ymmv. hence I suggest you to open 2 login simultaneously.

Chalipa · March 25

I'm also having lots of problems with my vps storage with servarica.

Going down and up constantly almost every day, and the support is not very helpful too. When contacting their support, they reply after a few hours simply saying it's responding to ping and there is no problem with the server.

@servarica_hani hopefully can do something about this.

gremeyer · March 25

@kode9 said:
To these yabs results seem ok to you guys ?

The iperf send results make it seem like one or more users are saturating the port so much that the host can't accept more connections.

jperkins · March 25

@gremeyer said:

@kode9 said:
To these yabs results seem ok to you guys ?

The iperf send results make it seem like one or more users are saturating the port so much that the host can't accept more connections.

Add this to your iperf command where n equals up to 8 and you may realize it isnt the port that is saturated, but possibly the thread that is limited.

-P, --parallel n
number of parallel client threads to run

tall_ice · March 26

@Chalipa said:
I'm also having lots of problems with my vps storage with servarica.

Going down and up constantly almost every day, and the support is not very helpful too. When contacting their support, they reply after a few hours simply saying it's responding to ping and there is no problem with the server.

@servarica_hani hopefully can do something about this.

While this is a new issue for me, I am definitely feeling frustrated too.

tall_ice · March 26

@kode9 said:

Block Size 4k (IOPS) 64k (IOPS)

Read 4.36 MB/s (1.0k) 7.32 MB/s (114)

Write 4.38 MB/s (1.0k) 7.73 MB/s (120)

Total 8.74 MB/s (2.1k) 15.05 MB/s (234)

Block Size 512k (IOPS) 1m (IOPS)

------ --- ---- ---- ----

Read 581.00 KB/s (1) 6.34 MB/s (6)

Write 632.00 KB/s (1) 6.45 MB/s (6)

Total 1.21 MB/s (2) 12.80 MB/s (12)

iperf3 Network Speed Tests (IPv4):

Provider | Location (Link) | Send Speed | Recv Speed | Ping
----- | ----- | ---- | ---- | ----
Clouvider | London, UK (10G) | 3.96 Mbits/sec | 1.44 Gbits/sec | 80.0 ms
Eranium | Amsterdam, NL (100G) | 4.26 Mbits/sec | 2.16 Gbits/sec | 83.2 ms
Uztelecom | Tashkent, UZ (10G) | 3.62 Mbits/sec | 826 Mbits/sec | 186 ms
Leaseweb | Singapore, SG (10G) | 2.11 Mbits/sec | 672 Mbits/sec | 244 ms
Clouvider | Los Angeles, CA, US (10G) | 5.58 Mbits/sec | 1.89 Gbits/sec | 73.8 ms
Leaseweb | NYC, NY, US (10G) | 31.7 Mbits/sec | 7.40 Gbits/sec | 9.24 ms
Performing IPv4 iperf3 recv test from Edgoo (Attempt #1 of 3)...

PS: that's all that I got `till it went down again.

woah. that's bad. One tip I can share for these unstable connection I have is use mosh.

tongtong · March 26

Servarica does have the problem of uneven resource scheduling of Xen, which can sometimes be fixed by restarting in the panel. Good support service will still make me choose it.

tall_ice · March 26

@tongtong said:
Servarica does have the problem of uneven resource scheduling of Xen, which can sometimes be fixed by restarting in the panel. Good support service will still make me choose it.

I too appreciate the good support service, but not having good access to the storage (the service I purchased) kind of a real issue.

tall_ice · March 26

After my switching server, my connection improved enough.

Before

Block Size | 4k            (IOPS) | 64k           (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 8.84 MB/s     (2.2k) | 22.85 MB/s     (357)
Write      | 8.88 MB/s     (2.2k) | 23.29 MB/s     (363)
Total      | 17.72 MB/s    (4.4k) | 46.14 MB/s     (720)
           |                      |                     
Block Size | 512k          (IOPS) | 1m            (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 35.60 MB/s      (69) | 141.00 KB/s      (0)
Write      | 37.38 MB/s      (73) | 187.00 KB/s      (0)
Total      | 72.98 MB/s     (142) | 328.00 KB/s      (0)

After

fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/mapper/debian-root):
---------------------------------
Block Size | 4k            (IOPS) | 64k           (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 58.67 MB/s   (14.6k) | 252.76 MB/s   (3.9k)
Write      | 58.78 MB/s   (14.6k) | 254.09 MB/s   (3.9k)
Total      | 117.46 MB/s  (29.3k) | 506.86 MB/s   (7.9k)
           |                      |                     
Block Size | 512k          (IOPS) | 1m            (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 233.33 MB/s    (455) | 225.33 MB/s    (220)
Write      | 245.73 MB/s    (479) | 240.33 MB/s    (234)
Total      | 479.07 MB/s    (934) | 465.67 MB/s    (454)

tall_ice · March 26

@kode9 You shoulld reach out to support. I hope your issue gets resolved!

terminator · March 26

@tall_ice said:
After my switching server, my connection improved enough.

Before

Block Size | 4k            (IOPS) | 64k           (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 8.84 MB/s     (2.2k) | 22.85 MB/s     (357)
Write      | 8.88 MB/s     (2.2k) | 23.29 MB/s     (363)
Total      | 17.72 MB/s    (4.4k) | 46.14 MB/s     (720)
           |                      |                     
Block Size | 512k          (IOPS) | 1m            (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 35.60 MB/s      (69) | 141.00 KB/s      (0)
Write      | 37.38 MB/s      (73) | 187.00 KB/s      (0)
Total      | 72.98 MB/s     (142) | 328.00 KB/s      (0)

After

fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/mapper/debian-root):
---------------------------------
Block Size | 4k            (IOPS) | 64k           (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 58.67 MB/s   (14.6k) | 252.76 MB/s   (3.9k)
Write      | 58.78 MB/s   (14.6k) | 254.09 MB/s   (3.9k)
Total      | 117.46 MB/s  (29.3k) | 506.86 MB/s   (7.9k)
           |                      |                     
Block Size | 512k          (IOPS) | 1m            (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 233.33 MB/s    (455) | 225.33 MB/s    (220)
Write      | 245.73 MB/s    (479) | 240.33 MB/s    (234)
Total      | 479.07 MB/s    (934) | 465.67 MB/s    (454)

Is this a permanent solution? I doubt it, this node will soon be overcrowded as well.

laey · March 26

@tenji said:
sorry if this not applied to you, but for me too many login bruteforcer on my ssh causing my ssh sessions so slow 😅
never experience overcrowded node, just noisy neighbour, with top command reveal steal above 10% enough to cause slow ssh response, ymmv.

That's true, it has happened to me before. I also set up Fail2Ban and configured SSH to be accessible only with an SSH key, disabling password login (when I used a DigitalOcean VPS).

By the way, all three of my dedicated servers from servarica is okay.

tall_ice · March 26

@terminator said:

@tall_ice said:
After my switching server, my connection improved enough.

Before

Block Size | 4k            (IOPS) | 64k           (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 8.84 MB/s     (2.2k) | 22.85 MB/s     (357)
Write      | 8.88 MB/s     (2.2k) | 23.29 MB/s     (363)
Total      | 17.72 MB/s    (4.4k) | 46.14 MB/s     (720)
           |                      |                     
Block Size | 512k          (IOPS) | 1m            (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 35.60 MB/s      (69) | 141.00 KB/s      (0)
Write      | 37.38 MB/s      (73) | 187.00 KB/s      (0)
Total      | 72.98 MB/s     (142) | 328.00 KB/s      (0)

After

fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/mapper/debian-root):
---------------------------------
Block Size | 4k            (IOPS) | 64k           (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 58.67 MB/s   (14.6k) | 252.76 MB/s   (3.9k)
Write      | 58.78 MB/s   (14.6k) | 254.09 MB/s   (3.9k)
Total      | 117.46 MB/s  (29.3k) | 506.86 MB/s   (7.9k)
           |                      |                     
Block Size | 512k          (IOPS) | 1m            (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 233.33 MB/s    (455) | 225.33 MB/s    (220)
Write      | 245.73 MB/s    (479) | 240.33 MB/s    (234)
Total      | 479.07 MB/s    (934) | 465.67 MB/s    (454)

Is this a permanent solution? I doubt it, this node will soon be overcrowded as well.

Well I can't really say if it's a permanent solution. I can only say it's okay right now. I can tolerate it if it's once a year or something. So far it's been once last month and once now. I can't comment on the future. Other longer term customer can comment. I hope servarica can improve on this. But for now I will hold.

@laey said:

@tenji said:
sorry if this not applied to you, but for me too many login bruteforcer on my ssh causing my ssh sessions so slow 😅
never experience overcrowded node, just noisy neighbour, with top command reveal steal above 10% enough to cause slow ssh response, ymmv.

That's true, it has happened to me before. I also set up Fail2Ban and configured SSH to be accessible only with an SSH key, disabling password login (when I used a DigitalOcean VPS).

By the way, all three of my dedicated servers from servarica is okay.

I am on shared. Glad to hear that dedicated is good. You paid for it!

JasonM · March 26

@tenji said: sorry if this not applied to you, but for me too many login bruteforcer on my ssh causing my ssh sessions so slow

yes, today its feels slow.
otherwise its good and fast.

Chalipa · March 26

I'm still having the same issue

Going down for a few mins then coming back online, after denying that they were having issue with the node, now they finally agreed to changed my vps to a new node. Moving backup data is a big headache for me. I don't know what's causing this and I'm not sure moving to a new node will resolve this issue.

yabs

root@castorage:~# curl -sL yabs.sh | bash
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
#              Yet-Another-Bench-Script              #
#                     v2025-01-01                    #
# https://github.com/masonr/yet-another-bench-script #
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #

Tue Mar 25 11:06:16 PM CET 2025

Basic System Information:
---------------------------------
Uptime     : 14 days, 15 hours, 4 minutes
Processor  : Intel(R) Xeon(R) Gold 6132 CPU @ 2.60GHz
CPU cores  : 2 @ 2600.033 MHz
AES-NI     :  Enabled
VM-x/AMD-V :  Disabled
RAM        : 1.9 GiB
Swap       : 1.9 GiB
Disk       : 1015.9 GiB
Distro     : Debian GNU/Linux 12 (bookworm)
Kernel     : 6.1.0-31-amd64
VM Type    : XEN
IPv4/IPv6  :  Online /  Offline

IPv4 Network Information:
---------------------------------
ISP        : Rica Web Services
ASN        : AS26832 Rica Web Services
Host       : Rica Web Services
Location   : Montreal, Quebec (QC)
Country    : Canada

fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/mapper/debian-root):
---------------------------------
Block Size | 4k            (IOPS) | 64k           (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 61.78 MB/s   (15.4k) | 212.39 MB/s   (3.3k)
Write      | 61.90 MB/s   (15.4k) | 213.51 MB/s   (3.3k)
Total      | 123.69 MB/s  (30.9k) | 425.90 MB/s   (6.6k)
           |                      |                     
Block Size | 512k          (IOPS) | 1m            (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 235.21 MB/s    (459) | 241.81 MB/s    (236)
Write      | 247.71 MB/s    (483) | 257.92 MB/s    (251)
Total      | 482.93 MB/s    (942) | 499.73 MB/s    (487)

iperf3 Network Speed Tests (IPv4):
---------------------------------
Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping           
-----           | -----                     | ----            | ----            | ----           
Clouvider       | London, UK (10G)          | 896 Mbits/sec   | 2.17 Gbits/sec  | 93.1 ms        
Eranium         | Amsterdam, NL (100G)      | 932 Mbits/sec   | 2.33 Gbits/sec  | 80.1 ms        
Uztelecom       | Tashkent, UZ (10G)        | 878 Mbits/sec   | 585 Mbits/sec   | 169 ms         
Leaseweb        | Singapore, SG (10G)       | 637 Mbits/sec   | 661 Mbits/sec   | 255 ms         
Clouvider       | Los Angeles, CA, US (10G) | 892 Mbits/sec   | 2.54 Gbits/sec  | 73.8 ms        
Leaseweb        | NYC, NY, US (10G)         | 1.01 Gbits/sec  | 6.77 Gbits/sec  | 9.68 ms        
Edgoo           | Sao Paulo, BR (1G)        | 867 Mbits/sec   | 1.05 Gbits/sec  | 133 ms         

Geekbench 6 Benchmark Test:
---------------------------------
Test            | Value                         
                |                               
Single Core     | 1015                          
Multi Core      | 1653                          
Full Test       | https://browser.geekbench.com/v6/cpu/11209847

YABS completed in 15 min 0 sec

kode9 · March 27

@tall_ice said:
woah. that's bad. One tip I can share for these unstable connection I have is use mosh.

I could. The only issue with that IS that I'm using the VM for rsync to another VM and I have a active connection to it 24/7 from my phone. So when the VM goes down, I could use SUPER MOSH it will still not help me upload stuff from my phone to the VM as long as the VM is down ...

Ziny · March 27

@tall_ice said:
Servarica down? Or extremely slow connection? or just me

your network ,freezing?

Howdy, Stranger!

Categories

In this Discussion

Servarica down?

Comments

## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ##

Yet-Another-Bench-Script

v2025-01-01

https://github.com/masonr/yet-another-bench-script

## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ##

Basic System Information:

IPv6 Network Information:

fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/mapper/ubuntu--vg-lv--root):

iperf3 Network Speed Tests (IPv4):

iperf3 Network Speed Tests (IPv4):

Howdy, Stranger!

Quick Links

Categories

In this Discussion

Servarica down?

Comments

## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ##

Yet-Another-Bench-Script

v2025-01-01

https://github.com/masonr/yet-another-bench-script

## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ##

Basic System Information:

IPv6 Network Information:

fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/mapper/ubuntu--vg-lv--root):

iperf3 Network Speed Tests (IPv4):

iperf3 Network Speed Tests (IPv4):