Need an explanation on this connection issue

Start from traceroute and measuring on which hop latency start to increase, and then checking modem SNR/error rate/etc in case it is hop between modem and ISP.

That hops where latency increase is a culprit. If they belong to ISP - then it is ISP to blame and it is their responsibility to fix their shit or explain why you are throttled.
Customer usually first hops, if first and second hop fine and it is traditional "single modem/router" network - your network is ok and thats proof of it.
Most of network engineers able to segment traceroute hop by IPs to customer, ISP network (sometimes visible different parts of it) and their backbone, to identify bottleneck.

For THGV it is tricky. Some hops inside THGV have latency because ICMP throttled. ping 10.0.25.10, thats internal host that should be bypassed.
Are you DSL or wireless/reseller?

ping to 10.0.25.10 should not be "busted on their end", but let's assume its still possible. thats internal path and it is bypassed from bandwidth policy.

Few more important questions:
1)Is it behaving different at early morning?
2)Did you check your ADSL modem stats? What are up/down speeds?
3)Can you try to ping 192.168.99.1 ? This is internal IP of THGV LNS server, so internal path totally ruled out.

I need to know answers to those questions, so i can know if any of following true:
1)Problem between Ogero(centrale LAC) and THGV equipment. Likely not, but i will ask if there is recently complaints about same issue
2)Problem with termination server bandwidth policy
3)Problem is on your side

That means unlikely it is THGV issue, as most of issues is load related and day dependent, also i just did checkup of all their systems and discussed with their techies, they said such complaints(ping goes higher during the load) they heard, but less than 5 customers, so it might be internal Ogero network related, which Ogero unlikely will bother to fix this days, but i am not entirely sure it is Ogero.
But definitely not THGV, as both DSL servers i checked and they are working fine, and ping 192.168.99.1 there is only one unit involved - THGV DSL server, and no other infrastructure.

As last resort, did you tried to use ethernet cable to modem? Sometimes it is just buggy modems, and on wifi path NAT offload doesnt work properly.

I've cleaned up the posts that got heated and will keep the topic running. Troubleshooting is always a gray area, it's generally difficult for any one person to be conclusively right without having the full picture/details.

This topic may help others going through the same thing.

Thank you all for pitching in!

xterm, it is not gray, it is tech, and troubleshooting is trivial, when customer give proper answers and have basic tech knowledge. I did created guidelines for support for such cases and did such troubleshooting myself endless number of times.
Also, let me clarify, i created all tech infrastructure on TH, and i am still running it, but sure not alone, with their team.
In this case it is clear that _one_ person can indeed identify issue and if it is in his zone of responsibility - fix it or provide detailed info that it will be fixed by third party (Ogero, but i can assure - they wont bulge doing that for one end-user).

For example:

newnetdb ~ # mtr 8.8.8.8 --report --report-cycles=10
Start: Tue Aug 2 12:54:28 2022
HOST: newnetdb Loss% Snt Last Avg Best Wrst StDev
1.|-- 192.168.1.1 0.0% 10 a b c d
2.|-- 192.168.99.1 0.0% 10 ....
3.|-- 10.0.253.x 0.0% 10 ....
4.|-- 10.1.16.1 0.0% 10 4547. 2832. 1500. 4547. 1216.0
5.|-- 172.16.79.229 0.0% 10 0.2 0.2 0.2 0.2 0.0
6.|-- ??? 100.0 10 0.0 0.0 0.0 0.0 0.0
7.|-- 172.16.49.1 0.0% 10 1.2 1.2 1.1 1.8 0.0
8.|-- 213.242.116.233 0.0% 10 37.7 43.5 37.5 80.1 13.8
9.|-- 72.14.212.254 0.0% 10 37.6 37.6 37.5 37.8 0.0
10.|-- 74.125.244.209 0.0% 10 37.8 37.7 37.7 37.8 0.0
11.|-- 142.250.46.95 0.0% 10 38.7 38.9 38.2 42.1 1.1
10.|-- dns.google 0.0% 10 37.6 37.6 37.6 37.7 0.0

Some hops i imitated, cause i dont have access to identical DSL account.
But it is obvious, thru monitoring traceroute, if latency start to jump on first hop - it is user wifi router.
On second hop it is usually ISP BRAS, if latency start to jitters there - it means either Ogero DSP issues, or ISP have problems on his BRAS, OR in some cases shaper issues on BRAS.
Then you can see those high values on hop 4 (probably also 3 have them), but they are normal if next hops have lower latency, it just means that router control-plane throttling ICMP, but what is important that latency on last hops is OK and such values can be ignored.
Sure i can provide references and proof for each word i did said.

Structure of Ogero network is approx like this

And latency visibly increasing on LNS point, which is not easy to troubleshoot at all. Likely it is congestion at Ogero centrale, especially as user admit that it happened with him once on another ISP (and obvious logic, what is common in this case? Ogero infra)

But in current case user is withholding proper data (traceroute) and making ugly remarks before even troubleshooting completed.
I'm putting my nose in this issue and trying to solve his problem as my personal initiative, and I voluntarily donate my time, going to all extent explaining how things work, and instead of continuing, he jump abruptly to conclusions and make such accusational comments:

I guess THGV is simply cutting corners due to the economic crisis in order to increase profit, as all these problems just started happening out of nowhere after the new internet prices came out, and it seems like anyone related to this company is always either blaming Ogero or the customer for their own problems.

So, in a situation where a person is behaving disrespectfully - would you waste time on him?

P.S. Please dont delete this post, i wrote it respectfully as i can and i believe it doesnt violate any rules and also have learning value, as if any tech trying to help and go out of his duty responsibilities, user should have some respect as well to him and his company.

nuclearcat wrote Ogero have very limited manpower, so their support covering most basic cases only.
Also their demarcation point is box in building, means at best (you might have to wait weeks for that) they will come and check line before box, and if it is ok - you are on your own.
If your modem lost settings? They wont bother telling how to put vci/vpi, the only your choice is to come to their office and they will program modem for you. AFAIK they dont even give you your own login/password.
Some website fail to open due routing issues? Good luck troubleshooting that with Ogero as end-user.

That makes my point even more true cause if i am with a private ISP using Ogeros dsl infrastructure if any problem occurs you will have 2 parties to deal with it will take a week or 2 for the ISP to figure out its a line problem for example and then a month for ogero to come fix it so why not cut corners? that is 1 more reason to go with ogero if they are your dsp

All ISP have onsite support, but it cost way more than account itself, literally nobody pay for it, even those who say they are ready to pay, when engineer arrive and fix their issue, they often refuse to pay (even was agreed otherwise), so most stopped offering it. If old man dont want to put vpi/vci, he can do easiest thing - bring his modem to ISP office.

so what makes them better than ogero? again nothing, ogero has support teams and Centrale all over lebanon but this man is like 2 hour drive from the ISP if the problem was with ogero it would have been a 10minute trip to fix it, but he was without internet for weeks for simple stuff.
If it costs more than the account to fix an issue then why provide services all over lebanon if you can provide support in one area? simply because it generates way more money as you are using ogero's infrastructure and paying them a fee so u can have as many accounts as u like without bothering with infrastructure other than couple of routers and servers and share the bandwidth between users this explains the tens of "ISPs" in this very small market

8-9 internal hops is quite normal, especially in Lebanon due government monopoly limitations (government force to split each connection to ISP/DSP + they dont allow anybody except ogero to own fiber and etc).
When you hide all those hops behind MPLS, it will look nicer, but much worse to troubleshoot.
Also, this is for example comcast, US:

OUTER:
traceroute to 216.58.217.142 (216.58.217.142), 64 hops max, 72 byte packets
1 router.asus.com (192.168.1.1) 4.018 ms 0.718 ms 0.655 ms
2 * * c-68-81-70-1.hsd1.pa.comcast.net (68.81.70.1) 4228.615 ms <————— UH YA, THAT’S NOT GOOD
3 68.85.194.209 (68.85.194.209) 19.755 ms 18.592 ms 10.102 ms
4 be-20-ar03.newcastle.de.panjde.comcast.net (69.139.192.153) 19.453 ms 33.912 ms 109.620 ms
5 be-33287-cr02.ashburn.va.ibone.comcast.net (68.86.94.249) 1736.663 ms 4787.462 ms 20.046 ms
6 he-0-10-0-0-pe04.ashburn.va.ibone.comcast.net (68.86.82.210) 16.548 ms 20.248 ms 19.318 ms
7 as36040-2-c.ashburn.va.ibone.comcast.net (75.149.229.86) 14.933 ms 24.071 ms 14.382 ms

By internal i mean within the ISP not in lebanon all IPs were in the same subnet
i know it is normal in lebanon to have lops of internal hops due to the situation data goes around in a circle from ogero to isp or maybe another dsp then ogero then international lol
in US the hops are across several regions not within same building like the isp i am referring to

DNA wrote That makes my point even more true cause if i am with a private ISP using Ogeros dsl infrastructure if any problem occurs you will have 2 parties to deal with it will take a week or 2 for the ISP to figure out its a line problem for example and then a month for ogero to come fix it so why not cut corners? that is 1 more reason to go with ogero if they are your dsp

I'm pointing that Ogero wont bother about your tricky backbone issues, like some app not working, or routing issue of website, if you are end-user.
Private ISP on high-class service type will do that and fast enough.

DNA wrote so what makes them better than ogero? again nothing, ogero has support teams and Centrale all over lebanon but this man is like 2 hour drive from the ISP if the problem was with ogero it would have been a 10minute trip to fix it, but he was without internet for weeks for simple stuff.
If it costs more than the account to fix an issue then why provide services all over lebanon if you can provide support in one area? simply because it generates way more money as you are using ogero's infrastructure and paying them a fee so u can have as many accounts as u like without bothering with infrastructure other than couple of routers and servers and share the bandwidth between users this explains the tens of "ISPs" in this very small market

Ogero have support teams that handle traditional telephone stuff, half-related to most common internet issues - broken wiring from box to centrale, sometimes they can try to solve bad noisy pairs(but that one as i see by feedback of their users - after quite lot of nagging). As soon as issue different - good luck.
You cannot bring your modem to ANY centrale. AFAIK you need to bring to Ogero central offices in Beirut.
Why provide services all over lebanon and not having local support offices? Just because average Lebanese citizen cannot afford paying for such level of service that ISP will have support office at walk distance. It means there is support or not, running offices all across country will end up in ARPU, your account cost.
And it is not just "few servers". Are you aware that Google, Facebook, Akamai, Netflix and many others running part of their offloading infrastructure at private ISPs? And private ISPs give priority on such traffic and sometimes even dont count it as FUP/quota? Also often ISP trying to run QoS, so your conferencing software will run, even if your area/connection is congested (depends on ISP and service type). On Ogero... i know many people switched to private companies, especially when remote education was only option, just because Ogero set too small upload on DSL or centrale is congested and they dont have any QoS set for users and they are unable to solve issue.

DNA wrote By internal i mean within the ISP not in lebanon all IPs were in the same subnet
i know it is normal in lebanon to have lops of internal hops due to the situation data goes around in a circle from ogero to isp or maybe another dsp then ogero then international lol
in US the hops are across several regions not within same building like the isp i am referring to

Thats just not correct understanding of ISP infrastructure.
US ISP might have infrastructure stretched all across the country, just because of country size.
ISP in Lebanon might have all hops in reserved subnets (10/8, 172.16/12, etc) cause they are not wasting real IPs, but that doesn't means it is one place, or stretched all across the country.
You cannot judge ISP infrastructure by traceroute unless you know exactly where each hop are and what it is doing.
Assuming anything just by numbers of hops or subnets - is not proper way at all.
I have some ISPs where DDoS protection scheme generate 1-2 hops with ips 10.x.x.x which are same as previous ones, but they are located abroad, in IX locations.

man u are mixing things up, u are talking about ogero as if most of the users and ISPs arent using their dsl to begin with, if ogero's upload is small switching to any other isp results in worse performance because Ogero is your DSP and upload will stay the same, i never said dont go with private ISPs when they can provide u service using their own infrastructure, but from my experience ogero's internet is the most stable its actually great no issues whatsoever i see no point in switching to private when i am connected to Ogeros DSLAM unless i am going microwave 4G or something

btw microwave is great for coorporate users across the country i mean without it its almost impossible to get anything done with ogero in most regions. all corporate users i work with have private isps and i even recommend this route, their support is better GDS support is very good also.

but as home users they are worse than ogero they are using Ogero's infrastructure to reach the whole country and make a buck yet they don't provide support and can easily blame ogero thats how its going, u can't provide a service with no support, but they are getting away with it stop denying that fact man and no i do not have to pay u to come fix my internet, its part of your business

That reminded me of our "ishtirak" guy if his generator broke he used to include on the bill "فرق صيانة" lol
like running a business should be 100% profit no costs or anything.

regarding modem configuration no they offer the service not only in beirut and in some cases they will ship it or something at least in my area u can leave the modem and take it after a week or so.( i am not sure in the current situation though).

And it is not just "few servers". Are you aware that Google, Facebook, Akamai, Netflix and many others running part of their offloading infrastructure at private ISPs?

yes i am aware of that,but still their internet service is couple of routers and brases, u do not have a backbone in the country you are simply using someone else's, yea yea you can add 100 servers for caching offloading or whatever but that doesn't change the fact that u can run an ISP in lebanon with a couple bgp router for your ASN couple other routers and a bras or 2.

Thats just not correct understanding of ISP infrastructure.
US ISP might have infrastructure stretched all across the country, just because of country size.
ISP in Lebanon might have all hops in reserved subnets (10/8, 172.16/12, etc) cause they are not wasting real IPs, but that doesn't means it is one place, or stretched all across the country.
You cannot judge ISP infrastructure by traceroute unless you know exactly where each hop are and what it is doing.
Assuming anything just by numbers of hops or subnets - is not proper way at all.
I have some ISPs where DDoS protection scheme generate 1-2 hops with ips 10.x.x.x which are same as previous ones, but they are located abroad, in IX locations.

i will not pretend to know how they do their things but seeing example 172.16.45.3 --> 172.16.45.5--> 172.16.46.1 etc... and all with similar ping meaning close proximity, feels so weird and unnecessary, but what do i know.