HAPPY NEW YEAR 2025 - MEGATHREAD

plumberg · December 2024

@emgh anyone bought yr soul or something yet?
Or still up for grabs

plumberg · December 2024

@ralf said:

@Blembim said:
maybe i'm not in a linux enterprise field, but can someone explain me why some company prefer Commercial Distro like RHEL or SUSE. they got special support or something like that?

This is going to sound as facetious as my comments about *aaS, but this is actually trying to be serious.

In general, the bigger company is, the more the likely it is that having someone to blame and shout at when something goes wrong is more important than not having any problems in the first place.

Stated differently, large companies generally would prefer something that works 99% of the time and a support contract to fix things for that 1% than a product that's so far worked 100% but doesn't offer any support contract in case it does fail. Obviously, the ideal is to be in that 100% bucket whilst also offering a support contract, but the occasional failure reminds the customer that the support is valuable. Perversely, there's also often a perception that when things break, it shows that they're complicated and so the people maintaining them are even more valuable.

RedHat was one of the earliest model for charging for Linux support, although originally it was literally just support and all their custom changes were also available for free. There was quite a big stink at the time when they shifted to RHEL for enterprise features that they didn't release in the free RedHat, and ultimately that led to their original distribution dying out after CentOS took base RedHat and back-ported almost everything from RHEL to it.

In my experience, most small companies would go with something like CentOS and try to do deal with any problems in-house with their "linux guy" because they've already accounted for paying him in the budget but a support contract is additional expense, but bigger companies take a different view - they've got budget for a certain number of developers because they're needed for feature work or whatever. They have an IT department to deal with provisioning servers and dealing with problems, and they in turn are usually worked pretty hard and so will often pay for support contracts to make their lives easier and can just add it to their budget saying it's a necessary expense, but also when stuff fails, the pressure isn't all on them personally as they can say "yeah, we've got blah company (who are THE EXPERTS) looking at it" and absolve themselves of the responsibilty.

Very well put. People don't realize the value of support until the day they need it.
All contracts I sign with vendors I make sure there is atleast a minimum basic support provided or mission critical as need be

Makes it easy to continue keeping the lights on.

Saragoldfarb · December 2024

@plumberg said:
@emgh anyone bought yr soul or something yet?
Or still up for grabs

His soul has not been reaped yet. I own his ass though.

plumberg · December 2024

@ehab said:

@ralf said:
This is going to sound as facetious as my comments about *aaS, but this is actually trying to be serious.

In general, the bigger company is, the more the likely it is that having someone to blame and shout at when something goes wrong is more important than not having any problems in the first place.

Stated differently, large companies generally would prefer something that works 99% of the time and a support contract to fix things for that 1% than a product that's so far worked 100% but doesn't offer any support contract in case it does fail. Obviously, the ideal is to be in that 100% bucket whilst also offering a support contract, but the occasional failure reminds the customer that the support is valuable. Perversely, there's also often a perception that when things break, it shows that they're complicated and so the people maintaining them are even more valuable.

RedHat was one of the earliest model for charging for Linux support, although originally it was literally just support and all their custom changes were also available for free. There was quite a big stink at the time when they shifted to RHEL for enterprise features that they didn't release in the free RedHat, and ultimately that led to their original distribution dying out after CentOS took base RedHat and back-ported almost everything from RHEL to it.

In my experience, most small companies would go with something like CentOS and try to do deal with any problems in-house with their "linux guy" because they've already accounted for paying him in the budget but a support contract is additional expense, but bigger companies take a different view - they've got budget for a certain number of developers because they're needed for feature work or whatever. They have an IT department to deal with provisioning servers and dealing with problems, and they in turn are usually worked pretty hard and so will often pay for support contracts to make their lives easier and can just add it to their budget saying it's a necessary expense, but also when stuff fails, the pressure isn't all on them personally as they can say "yeah, we've got blah company (who are THE EXPERTS) looking at it" and absolve themselves of the responsibilty.

i can't read all that

Yeah
@ralf reminds me of the continuous long wall of texts by @VirMach

plumberg · December 2024

@ralf said:

@ehab said:

@ralf said:
This is going to sound as facetious as my comments about *aaS, but this is actually trying to be serious.

In general, the bigger company is, the more the likely it is that having someone to blame and shout at when something goes wrong is more important than not having any problems in the first place.

Stated differently, large companies generally would prefer something that works 99% of the time and a support contract to fix things for that 1% than a product that's so far worked 100% but doesn't offer any support contract in case it does fail. Obviously, the ideal is to be in that 100% bucket whilst also offering a support contract, but the occasional failure reminds the customer that the support is valuable. Perversely, there's also often a perception that when things break, it shows that they're complicated and so the people maintaining them are even more valuable.

RedHat was one of the earliest model for charging for Linux support, although originally it was literally just support and all their custom changes were also available for free. There was quite a big stink at the time when they shifted to RHEL for enterprise features that they didn't release in the free RedHat, and ultimately that led to their original distribution dying out after CentOS took base RedHat and back-ported almost everything from RHEL to it.

In my experience, most small companies would go with something like CentOS and try to do deal with any problems in-house with their "linux guy" because they've already accounted for paying him in the budget but a support contract is additional expense, but bigger companies take a different view - they've got budget for a certain number of developers because they're needed for feature work or whatever. They have an IT department to deal with provisioning servers and dealing with problems, and they in turn are usually worked pretty hard and so will often pay for support contracts to make their lives easier and can just add it to their budget saying it's a necessary expense, but also when stuff fails, the pressure isn't all on them personally as they can say "yeah, we've got blah company (who are THE EXPERTS) looking at it" and absolve themselves of the responsibilty.

i can't read all that

TLDR:

Kickin in deez nuts solves all problems

plumberg · December 2024

MS said:
Came for deals, got walls of text.

Beauty of LET

cainyxues · December 2024

Me going guys if 3 usd deals available for transfer please pm
Regaurds

plumberg · December 2024

@Blembim said:

@ralf said:

@Blembim said:
maybe i'm not in a linux enterprise field, but can someone explain me why some company prefer Commercial Distro like RHEL or SUSE. they got special support or something like that?

This is going to sound as facetious as my comments about *aaS, but this is actually trying to be serious.

In general, the bigger company is, the more the likely it is that having someone to blame and shout at when something goes wrong is more important than not having any problems in the first place.

Stated differently, large companies generally would prefer something that works 99% of the time and a support contract to fix things for that 1% than a product that's so far worked 100% but doesn't offer any support contract in case it does fail. Obviously, the ideal is to be in that 100% bucket whilst also offering a support contract, but the occasional failure reminds the customer that the support is valuable. Perversely, there's also often a perception that when things break, it shows that they're complicated and so the people maintaining them are even more valuable.

RedHat was one of the earliest model for charging for Linux support, although originally it was literally just support and all their custom changes were also available for free. There was quite a big stink at the time when they shifted to RHEL for enterprise features that they didn't release in the free RedHat, and ultimately that led to their original distribution dying out after CentOS took base RedHat and back-ported almost everything from RHEL to it.

In my experience, most small companies would go with something like CentOS and try to do deal with any problems in-house with their "linux guy" because they've already accounted for paying him in the budget but a support contract is additional expense, but bigger companies take a different view - they've got budget for a certain number of developers because they're needed for feature work or whatever. They have an IT department to deal with provisioning servers and dealing with problems, and they in turn are usually worked pretty hard and so will often pay for support contracts to make their lives easier and can just add it to their budget saying it's a necessary expense, but also when stuff fails, the pressure isn't all on them personally as they can say "yeah, we've got blah company (who are THE EXPERTS) looking at it" and absolve themselves of the responsibilty.

So, it something like large company pay external contractor(Redhat etc) that take responsiblity when shit happens with their distro, plus support or custom features stuff?

Yes. To make sure someone is there to

put the blame
get some peace of mind that someone is there to help
cope

plumberg · December 2024

@cainyxues said:

@Blembim said:

So, it something like large company pay external contractor(Redhat etc) that take responsiblity when shit happens with their distro, plus support or custom features stuff?

In short, YES

In long?

plumberg · December 2024

@ehab said:

@cainyxues said:
Well those things don't get "fixed" easily so you are out of options

nice to meet your acquaintance

get ready

That looked really hard. .

plumberg · December 2024

@Decicus said:
Just took my hourly dose, anyone else need theirs?

I toned it down to every 3 hours. So yeah I am good for now

Saragoldfarb · December 2024

@plumberg said:

@ehab said:

@cainyxues said:
Well those things don't get "fixed" easily so you are out of options

nice to meet your acquaintance

get ready

That looked really hard. .

Regular Friday afternoon exercise.

plumberg · December 2024

@admax said:
I think I need to sleep to cope.

Sleeping is for n00bs
Stay awake
Just less than 48 hours to enjoy 2024.

admax · December 2024

@plumberg said:

@Blembim said:

@ralf said:

@Blembim said:
maybe i'm not in a linux enterprise field, but can someone explain me why some company prefer Commercial Distro like RHEL or SUSE. they got special support or something like that?

This is going to sound as facetious as my comments about *aaS, but this is actually trying to be serious.

In general, the bigger company is, the more the likely it is that having someone to blame and shout at when something goes wrong is more important than not having any problems in the first place.

Stated differently, large companies generally would prefer something that works 99% of the time and a support contract to fix things for that 1% than a product that's so far worked 100% but doesn't offer any support contract in case it does fail. Obviously, the ideal is to be in that 100% bucket whilst also offering a support contract, but the occasional failure reminds the customer that the support is valuable. Perversely, there's also often a perception that when things break, it shows that they're complicated and so the people maintaining them are even more valuable.

RedHat was one of the earliest model for charging for Linux support, although originally it was literally just support and all their custom changes were also available for free. There was quite a big stink at the time when they shifted to RHEL for enterprise features that they didn't release in the free RedHat, and ultimately that led to their original distribution dying out after CentOS took base RedHat and back-ported almost everything from RHEL to it.

In my experience, most small companies would go with something like CentOS and try to do deal with any problems in-house with their "linux guy" because they've already accounted for paying him in the budget but a support contract is additional expense, but bigger companies take a different view - they've got budget for a certain number of developers because they're needed for feature work or whatever. They have an IT department to deal with provisioning servers and dealing with problems, and they in turn are usually worked pretty hard and so will often pay for support contracts to make their lives easier and can just add it to their budget saying it's a necessary expense, but also when stuff fails, the pressure isn't all on them personally as they can say "yeah, we've got blah company (who are THE EXPERTS) looking at it" and absolve themselves of the responsibilty.

So, it something like large company pay external contractor(Redhat etc) that take responsiblity when shit happens with their distro, plus support or custom features stuff?

Yes. To make sure someone is there to

put the blame

get some peace of mind that someone is there to help

cope

Great»»»cope

plumberg · December 2024

@cainyxues said:
Epic games is running free games giveaway so whoever wants may go and collect them

I have been so out of touch with these games.
Last was Capture the flag/ CS/ road rash

After that everything seems alien to me

plumberg · December 2024

@Saragoldfarb said:

@plumberg said:
@emgh anyone bought yr soul or something yet?
Or still up for grabs

His soul has not been reaped yet. I own his ass though.

Awesum

ehab · December 2024

@Saragoldfarb

why don't u change into something else

Decicus · December 2024

@plumberg said:

@Decicus said:

@plumberg said:

@Decicus said:

@plumberg said:

@Decicus said:

@plumberg said:

@Decicus said:

@plumberg said:
Who was your first vps provider?

Not too proud , for me it was chicagovps

NFOservers back in 2013. In fact, I still have a server with them and I've had it for almost 9 years:

You signed up for this server on Friday, January 22, 2016

Wow. Seems they are a good provider.

They've been fairly reliable over the years for sure. Some blips here and there, but nothing crazy.

I should do a YABS of the server I still have with them one day. I believe it's still on HDD (in RAID1 or something), so disk speeds are probably not the greatest, but not sure.

Even big names get blips left and right
It's commendable they have still managed to stay alive. This market is so saturated, profit margins are razor thin.

Yeah would love to see the yabs

Oh yeah, for sure. They're usually quite on the ball when it comes to issues (even minor issues with upstreams).

Anyway, decided to check this server (and did some upgrades/reboot).

2 vCPU / 2 GB RAM / 200 GB disk / 12 TB @ whatever max speed they have (some nws.sh tests gave me around ~2 Gbps) for 8.99 USD per month.

Not IPv6-enabled apparently (I would have to contact support), but this is a mostly-idle server anyway so I don't really care

YABS - No idea what happened with receive speeds here tbh:

# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
#              Yet-Another-Bench-Script              #
#                     v2024-12-20                    #
# https://github.com/masonr/yet-another-bench-script #
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #

Mon Dec 30 11:58:00 AM EST 2024

Basic System Information:
---------------------------------
Uptime     : 0 days, 0 hours, 2 minutes
Processor  : AMD EPYC 7543 32-Core Processor
CPU cores  : 2 @ 2800.285 MHz
AES-NI     : ✔ Enabled
VM-x/AMD-V : ❌ Disabled
RAM        : 1.9 GiB
Swap       : 4.0 GiB
Disk       : 196.8 GiB
Distro     : Debian GNU/Linux 12 (bookworm)
Kernel     : 6.1.0-28-amd64
VM Type    : XEN
IPv4/IPv6  : ✔ Online / ❌ Offline

IPv4 Network Information:
---------------------------------
ISP        : Nuclearfallout Enterprises, Inc.
ASN        : AS14586 Nuclearfallout Enterprises, Inc.
Host       : Nuclearfallout Enterprises, Inc
Location   : Chicago, Illinois (IL)
Country    : United States

fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/xvda2):
---------------------------------
Block Size | 4k            (IOPS) | 64k           (IOPS)
  ------   | ---            ----  | ----           ----
Read       | 55.44 MB/s   (13.8k) | 230.25 MB/s   (3.5k)
Write      | 55.55 MB/s   (13.8k) | 231.46 MB/s   (3.6k)
Total      | 111.00 MB/s  (27.7k) | 461.72 MB/s   (7.2k)
           |                      |
Block Size | 512k          (IOPS) | 1m            (IOPS)
  ------   | ---            ----  | ----           ----
Read       | 475.19 MB/s    (928) | 604.21 MB/s    (590)
Write      | 500.44 MB/s    (977) | 644.45 MB/s    (629)
Total      | 975.64 MB/s   (1.9k) | 1.24 GB/s     (1.2k)

iperf3 Network Speed Tests (IPv4):
---------------------------------
Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping
-----           | -----                     | ----            | ----            | ----
Clouvider       | London, UK (10G)          | 1.21 Gbits/sec  | 23.8 Mbits/sec  | 85.2 ms
Eranium         | Amsterdam, NL (100G)      | 1.20 Gbits/sec  | 501 Mbits/sec   | 96.2 ms
Uztelecom       | Tashkent, UZ (10G)        | 877 Mbits/sec   | 23.8 Mbits/sec  | 182 ms
Leaseweb        | Singapore, SG (10G)       | 612 Mbits/sec   | busy            | 230 ms
Clouvider       | Los Angeles, CA, US (10G) | 1.76 Gbits/sec  | 74.4 Mbits/sec  | 47.3 ms
Leaseweb        | NYC, NY, US (10G)         | 2.46 Gbits/sec  | busy            | 17.1 ms
Edgoo           | Sao Paulo, BR (1G)        | 1.07 Gbits/sec  | 42.4 Mbits/sec  | 137 ms

Geekbench 6 Benchmark Test:
---------------------------------
Test            | Value
                |
Single Core     | 1381
Multi Core      | 2508
Full Test       | https://browser.geekbench.com/v6/cpu/9689047

YABS completed in 14 min 2 sec

nws.sh - Removed the FAILED tests to make it a bit smaller:

---------------------------------- nws.sh ---------------------------------
      A simple script to bench network performance using speedtest-cli
---------------------------------------------------------------------------
 Version            : v2024.12.20
 Global Speedtest   : wget -qO- nws.sh | bash
 Region Speedtest   : wget -qO- nws.sh | bash -s -- -r <region>
---------------------------------------------------------------------------
 Basic System Info
---------------------------------------------------------------------------
 CPU Model          : AMD EPYC 7543 32-Core Processor
 CPU Cores          : 2 @ 2800.285 MHz
 CPU Cache          : 512 KB
 AES-NI             : ✔ Enabled
 VM-x/AMD-V         : ❌ Disabled
 Total Disk         : 196.8 GB (119.4 GB Used)
 Total RAM          : 1.9 GB (473.9 MB Used)
 Total Swap         : 4.0 GB (135.5 MB Used)
 System uptime      : 0 days, 0 hour 18 min
 Load average       : 0.23, 0.68, 0.46
 OS                 : Debian GNU/Linux 12
 Arch               : x86_64 (64 Bit)
 Kernel             : 6.1.0-28-amd64
 Virtualization     : XEN
 TCP Control        :
---------------------------------------------------------------------------
 Basic Network Info
---------------------------------------------------------------------------
 IPv6 Access        : ❌ Offline
 IPv4 Access        : ✔ Online
 ISP                : Nuclearfallout Enterprises, Inc.
 ASN                : AS14586 Nuclearfallout Enterprises, Inc.
 Host               : Nuclearfallout Enterprises, Inc
 Location           : Chicago, Illinois-IL, United States
---------------------------------------------------------------------------
 Speedtest.net (Region: GLOBAL)
---------------------------------------------------------------------------
 Location         Latency     Loss    DL Speed       UP Speed       Server

 ISP: Nuclearfallout Enterprises

 Nearest          0.67 ms     0.0%    1249.16 Mbps   5500.22 Mbps   Nitel - Chicago, IL

 Kochi, IN        237.15 ms   0.0%    42.43 Mbps     367.85 Mbps    Asianet Broadband - Cochin
 Bangalore, IN    265.34 ms   0.0%    29.34 Mbps     316.91 Mbps    Bharti Airtel Ltd - Bangalore
 Chennai, IN      274.74 ms   0.0%    2.23 Mbps      249.82 Mbps    Jio - Chennai
 Mumbai, IN       201.87 ms   0.0%    11.02 Mbps     340.90 Mbps    Tata Teleservices Ltd - Mumbai
 Delhi, IN        285.79 ms   0.0%    77.60 Mbps     279.58 Mbps    Tata Play Fiber - New Delhi

 Seattle, US      49.37 ms    N/A     42.62 Mbps     994.13 Mbps    Comcast - Seattle, WA
 Los Angeles, US  42.21 ms    0.0%    39.83 Mbps     1205.30 Mbps   ReliableSite Hosting - Los Angeles, CA
 Dallas, US       22.36 ms    0.0%    67.94 Mbps     1199.43 Mbps   i3D.net - Dallas, TX
 Miami, US        34.25 ms    N/A     123.48 Mbps    828.98 Mbps    Dish Wireless - Miami, FL
 New York, US     16.64 ms    0.0%    1809.82 Mbps   2000.19 Mbps   GSL Networks - New York, NY

 London, UK       85.46 ms    0.0%    1713.54 Mbps   1049.41 Mbps   VeloxServ Communications - London
---------------------------------------------------------------------------
 Avg DL Speed       : 434.08 Mbps
 Avg UL Speed       : 1194.39 Mbps

 Total DL Data      : 6.97 GB
 Total UL Data      : 14.38 GB
 Total Data         : 21.35 GB
---------------------------------------------------------------------------
 Duration           : 7 min 0 sec
 System Time        : 30/12/2024 - 12:20:25 EST
 Total Script Runs  : 93681
---------------------------------------------------------------------------

Thnx

Seems a bit heavy on the wallet considering it's hdd.
But decent one overall

Yeah I don't think it's a great deal or anything, it's just I've had it for so long it feels weird to cancel it lol. Luckily it's "only" $8.99 per month, which... is not that much if I compare it to other servers in my spreadsheet.

Woah....

How many servers and avg price ?

94 servers

Actually make that 95, because I just grabbed one of those hourly restock specials from @ZachNuyek lmao

Overall average price is around 15 USD per month (so you can do the math :P), but there are a few important things to note:

There are some not-cheap dedis that heavily bump these numbers up.
Some providers charge me Norwegian VAT, which is 25%. Does not help with the final price, that's for sure.
Some of these servers are offset since I do some freelance work from time to time, where I also manage the server where things are hosted. For the servers that this is relevant (a few of the dedis), I do end up with a slight profit.

Do I have a problem? Absolutely.
Has it been worse? Yes. It used to be a good bit over 100 before.
Am I still going to grab deals that slows down the reduction of my spreadsheet? Yep. Gotta cope somehow.

emgh · December 2024

@plumberg said:

@emgh said:

@plumberg said:

I think the new trend now is doing a hybrid of cloud and in-house.

Lots of companies are feeling the heat of the ridiculous bills for being always available and being on the cloud.

Most of the time it was just hype which led teams make a decision to go 💯 cloud only to realize it's not that Kool.

In my last company, the data science teams were excited to move to the cloud, and they didn't realize how messed up the billing would be. Spun up a couple large ass 1tb ram clusters and forgot to terminate them after use. And a $ 6figure bill was slapped

I warned them of this but they are like, stay out bruddah. We know what we doing.

Was a nice.. I told you so moment...

I agree

Or at least ’in-house’ as in dedis with OVH/Hetzner, not always on-premise

I also love how the lack of basically any deals turned this into a general discussion thread

Do your magic bud. We are gonna enjoy your presence for a few hours only.

plumberg · December 2024

@Saragoldfarb said:

@plumberg said:

@ehab said:

@cainyxues said:
Well those things don't get "fixed" easily so you are out of options

nice to meet your acquaintance

get ready

That looked really hard. .

Regular Friday afternoon exercise.

Got some goose bumps seeing that and now your reply added

Back to every hour of coping dose.

emgh · December 2024

I wonder if we'll be able to beat this current ratio of 1 deal per 1 000 comments

plumberg · December 2024

@cainyxues said:
Me going guys if 3 usd deals available for transfer please pm
Regaurds

Regards

plumberg · December 2024

@admax said:

@plumberg said:

@Blembim said:

@ralf said:

@Blembim said:
maybe i'm not in a linux enterprise field, but can someone explain me why some company prefer Commercial Distro like RHEL or SUSE. they got special support or something like that?

This is going to sound as facetious as my comments about *aaS, but this is actually trying to be serious.

In general, the bigger company is, the more the likely it is that having someone to blame and shout at when something goes wrong is more important than not having any problems in the first place.

Stated differently, large companies generally would prefer something that works 99% of the time and a support contract to fix things for that 1% than a product that's so far worked 100% but doesn't offer any support contract in case it does fail. Obviously, the ideal is to be in that 100% bucket whilst also offering a support contract, but the occasional failure reminds the customer that the support is valuable. Perversely, there's also often a perception that when things break, it shows that they're complicated and so the people maintaining them are even more valuable.

RedHat was one of the earliest model for charging for Linux support, although originally it was literally just support and all their custom changes were also available for free. There was quite a big stink at the time when they shifted to RHEL for enterprise features that they didn't release in the free RedHat, and ultimately that led to their original distribution dying out after CentOS took base RedHat and back-ported almost everything from RHEL to it.

In my experience, most small companies would go with something like CentOS and try to do deal with any problems in-house with their "linux guy" because they've already accounted for paying him in the budget but a support contract is additional expense, but bigger companies take a different view - they've got budget for a certain number of developers because they're needed for feature work or whatever. They have an IT department to deal with provisioning servers and dealing with problems, and they in turn are usually worked pretty hard and so will often pay for support contracts to make their lives easier and can just add it to their budget saying it's a necessary expense, but also when stuff fails, the pressure isn't all on them personally as they can say "yeah, we've got blah company (who are THE EXPERTS) looking at it" and absolve themselves of the responsibilty.

So, it something like large company pay external contractor(Redhat etc) that take responsiblity when shit happens with their distro, plus support or custom features stuff?

Yes. To make sure someone is there to

put the blame

get some peace of mind that someone is there to help

cope

Great»»»cope

Regards

ehab · December 2024

@plumberg said:

@Saragoldfarb said:

Back to every hour of coping dose.

careful on weekdays she is more deceptive

i wouldn't be surprised if she does this

plumberg · December 2024

@emgh said:
I wonder if we'll be able to beat this current ratio of 1 deal per 1 000 comments

No ways man
No ways

plumberg · December 2024

@Decicus said:

@plumberg said:

@Decicus said:

@plumberg said:

@Decicus said:

@plumberg said:

@Decicus said:

@plumberg said:

@Decicus said:

@plumberg said:
Who was your first vps provider?

Not too proud , for me it was chicagovps

NFOservers back in 2013. In fact, I still have a server with them and I've had it for almost 9 years:

You signed up for this server on Friday, January 22, 2016

Wow. Seems they are a good provider.

They've been fairly reliable over the years for sure. Some blips here and there, but nothing crazy.

I should do a YABS of the server I still have with them one day. I believe it's still on HDD (in RAID1 or something), so disk speeds are probably not the greatest, but not sure.

Even big names get blips left and right
It's commendable they have still managed to stay alive. This market is so saturated, profit margins are razor thin.

Yeah would love to see the yabs

Oh yeah, for sure. They're usually quite on the ball when it comes to issues (even minor issues with upstreams).

Anyway, decided to check this server (and did some upgrades/reboot).

2 vCPU / 2 GB RAM / 200 GB disk / 12 TB @ whatever max speed they have (some nws.sh tests gave me around ~2 Gbps) for 8.99 USD per month.

Not IPv6-enabled apparently (I would have to contact support), but this is a mostly-idle server anyway so I don't really care

YABS - No idea what happened with receive speeds here tbh:

# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
#              Yet-Another-Bench-Script              #
#                     v2024-12-20                    #
# https://github.com/masonr/yet-another-bench-script #
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #

Mon Dec 30 11:58:00 AM EST 2024

Basic System Information:
---------------------------------
Uptime     : 0 days, 0 hours, 2 minutes
Processor  : AMD EPYC 7543 32-Core Processor
CPU cores  : 2 @ 2800.285 MHz
AES-NI     : ✔ Enabled
VM-x/AMD-V : ❌ Disabled
RAM        : 1.9 GiB
Swap       : 4.0 GiB
Disk       : 196.8 GiB
Distro     : Debian GNU/Linux 12 (bookworm)
Kernel     : 6.1.0-28-amd64
VM Type    : XEN
IPv4/IPv6  : ✔ Online / ❌ Offline

IPv4 Network Information:
---------------------------------
ISP        : Nuclearfallout Enterprises, Inc.
ASN        : AS14586 Nuclearfallout Enterprises, Inc.
Host       : Nuclearfallout Enterprises, Inc
Location   : Chicago, Illinois (IL)
Country    : United States

fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/xvda2):
---------------------------------
Block Size | 4k            (IOPS) | 64k           (IOPS)
  ------   | ---            ----  | ----           ----
Read       | 55.44 MB/s   (13.8k) | 230.25 MB/s   (3.5k)
Write      | 55.55 MB/s   (13.8k) | 231.46 MB/s   (3.6k)
Total      | 111.00 MB/s  (27.7k) | 461.72 MB/s   (7.2k)
           |                      |
Block Size | 512k          (IOPS) | 1m            (IOPS)
  ------   | ---            ----  | ----           ----
Read       | 475.19 MB/s    (928) | 604.21 MB/s    (590)
Write      | 500.44 MB/s    (977) | 644.45 MB/s    (629)
Total      | 975.64 MB/s   (1.9k) | 1.24 GB/s     (1.2k)

iperf3 Network Speed Tests (IPv4):
---------------------------------
Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping
-----           | -----                     | ----            | ----            | ----
Clouvider       | London, UK (10G)          | 1.21 Gbits/sec  | 23.8 Mbits/sec  | 85.2 ms
Eranium         | Amsterdam, NL (100G)      | 1.20 Gbits/sec  | 501 Mbits/sec   | 96.2 ms
Uztelecom       | Tashkent, UZ (10G)        | 877 Mbits/sec   | 23.8 Mbits/sec  | 182 ms
Leaseweb        | Singapore, SG (10G)       | 612 Mbits/sec   | busy            | 230 ms
Clouvider       | Los Angeles, CA, US (10G) | 1.76 Gbits/sec  | 74.4 Mbits/sec  | 47.3 ms
Leaseweb        | NYC, NY, US (10G)         | 2.46 Gbits/sec  | busy            | 17.1 ms
Edgoo           | Sao Paulo, BR (1G)        | 1.07 Gbits/sec  | 42.4 Mbits/sec  | 137 ms

Geekbench 6 Benchmark Test:
---------------------------------
Test            | Value
                |
Single Core     | 1381
Multi Core      | 2508
Full Test       | https://browser.geekbench.com/v6/cpu/9689047

YABS completed in 14 min 2 sec

nws.sh - Removed the FAILED tests to make it a bit smaller:

---------------------------------- nws.sh ---------------------------------
      A simple script to bench network performance using speedtest-cli
---------------------------------------------------------------------------
 Version            : v2024.12.20
 Global Speedtest   : wget -qO- nws.sh | bash
 Region Speedtest   : wget -qO- nws.sh | bash -s -- -r <region>
---------------------------------------------------------------------------
 Basic System Info
---------------------------------------------------------------------------
 CPU Model          : AMD EPYC 7543 32-Core Processor
 CPU Cores          : 2 @ 2800.285 MHz
 CPU Cache          : 512 KB
 AES-NI             : ✔ Enabled
 VM-x/AMD-V         : ❌ Disabled
 Total Disk         : 196.8 GB (119.4 GB Used)
 Total RAM          : 1.9 GB (473.9 MB Used)
 Total Swap         : 4.0 GB (135.5 MB Used)
 System uptime      : 0 days, 0 hour 18 min
 Load average       : 0.23, 0.68, 0.46
 OS                 : Debian GNU/Linux 12
 Arch               : x86_64 (64 Bit)
 Kernel             : 6.1.0-28-amd64
 Virtualization     : XEN
 TCP Control        :
---------------------------------------------------------------------------
 Basic Network Info
---------------------------------------------------------------------------
 IPv6 Access        : ❌ Offline
 IPv4 Access        : ✔ Online
 ISP                : Nuclearfallout Enterprises, Inc.
 ASN                : AS14586 Nuclearfallout Enterprises, Inc.
 Host               : Nuclearfallout Enterprises, Inc
 Location           : Chicago, Illinois-IL, United States
---------------------------------------------------------------------------
 Speedtest.net (Region: GLOBAL)
---------------------------------------------------------------------------
 Location         Latency     Loss    DL Speed       UP Speed       Server

 ISP: Nuclearfallout Enterprises

 Nearest          0.67 ms     0.0%    1249.16 Mbps   5500.22 Mbps   Nitel - Chicago, IL

 Kochi, IN        237.15 ms   0.0%    42.43 Mbps     367.85 Mbps    Asianet Broadband - Cochin
 Bangalore, IN    265.34 ms   0.0%    29.34 Mbps     316.91 Mbps    Bharti Airtel Ltd - Bangalore
 Chennai, IN      274.74 ms   0.0%    2.23 Mbps      249.82 Mbps    Jio - Chennai
 Mumbai, IN       201.87 ms   0.0%    11.02 Mbps     340.90 Mbps    Tata Teleservices Ltd - Mumbai
 Delhi, IN        285.79 ms   0.0%    77.60 Mbps     279.58 Mbps    Tata Play Fiber - New Delhi

 Seattle, US      49.37 ms    N/A     42.62 Mbps     994.13 Mbps    Comcast - Seattle, WA
 Los Angeles, US  42.21 ms    0.0%    39.83 Mbps     1205.30 Mbps   ReliableSite Hosting - Los Angeles, CA
 Dallas, US       22.36 ms    0.0%    67.94 Mbps     1199.43 Mbps   i3D.net - Dallas, TX
 Miami, US        34.25 ms    N/A     123.48 Mbps    828.98 Mbps    Dish Wireless - Miami, FL
 New York, US     16.64 ms    0.0%    1809.82 Mbps   2000.19 Mbps   GSL Networks - New York, NY

 London, UK       85.46 ms    0.0%    1713.54 Mbps   1049.41 Mbps   VeloxServ Communications - London
---------------------------------------------------------------------------
 Avg DL Speed       : 434.08 Mbps
 Avg UL Speed       : 1194.39 Mbps

 Total DL Data      : 6.97 GB
 Total UL Data      : 14.38 GB
 Total Data         : 21.35 GB
---------------------------------------------------------------------------
 Duration           : 7 min 0 sec
 System Time        : 30/12/2024 - 12:20:25 EST
 Total Script Runs  : 93681
---------------------------------------------------------------------------

Thnx

Seems a bit heavy on the wallet considering it's hdd.
But decent one overall

Yeah I don't think it's a great deal or anything, it's just I've had it for so long it feels weird to cancel it lol. Luckily it's "only" $8.99 per month, which... is not that much if I compare it to other servers in my spreadsheet.

Woah....

How many servers and avg price ?

94 servers

Actually make that 95, because I just grabbed one of those hourly restock specials from @ZachNuyek lmao

Overall average price is around 15 USD per month (so you can do the math :P), but there are a few important things to note:

There are some not-cheap dedis that heavily bump these numbers up.
Some providers charge me Norwegian VAT, which is 25%. Does not help with the final price, that's for sure.
Some of these servers are offset since I do some freelance work from time to time, where I also manage the server where things are hosted. For the servers that this is relevant (a few of the dedis), I do end up with a slight profit.

Do I have a problem? Absolutely.
Has it been worse? Yes. It used to be a good bit over 100 before.
Am I still going to grab deals that slows down the reduction of my spreadsheet? Yep. Gotta cope somehow.

As long as you get something back and if you feel you are in the green, it's all good.

plumberg · December 2024

@ehab said:

@plumberg said:

@Saragoldfarb said:

Back to every hour of coping dose.

careful on weekdays she is more deceptive

i wouldn't be surprised if she does this

Lol 😆 🤣 😂
Gotta be careful

plumberg · December 2024

Coping regards,

No deals

ehab · December 2024

@beanman109 i want you back the way i know you

is it no sleep made you fuzzy and can't see the mirror anymore?

plumberg · December 2024

@ehab how do you plan to enter the new year?

Howdy, Stranger!

Categories

In this Discussion

HAPPY NEW YEAR 2025 - MEGATHREAD

Comments

Great»»»cope

Great»»»cope

Howdy, Stranger!

Quick Links

Categories

In this Discussion

HAPPY NEW YEAR 2025 - MEGATHREAD

Comments

Great»»»cope

Great»»»cope