Jump to content

Arrexis

Member
  • Posts

    9
  • Joined

  • Last visited

Awards

This user doesn't have any awards

Arrexis's Achievements

  1. Regardless of the ping, I would grab the fiber connection. Not only is it going to offer better speeds and most likely stability, but it will also open you up to speed upgrades in the future (if your provider wants to). Heres to hoping they stop their bullshit and start offering actual speeds. Oh you mentioned QOS, contact the fiber ISP and ask them if the SOHO-router they supply will offer QOS. If not, I would reccomend you aquire your own router with QOS if the ISP allows that. In fact, I recommend using a non-ISP provided router to everyone, since the ISP provided ones are very often underpowered and lacking features. If you must use an ISP provided router/multimodem, see if you can put it in bridge mode and use your own router.
  2. So just to clarify, the 2 and 5 Mbps were just for comparison sake and is not the speeds offered by your provider? Technically there will be no significant difference in the speed they provide you, 5 Megabit is 5 Megabit. But fiber tends to have lower latency (ping/ms) and depending on your provider, better uptime/stability. Having a fiber line is a good thing, since it can supply vastly better speeds than DSL is capable of, and has better options of increasing your speed. Another thing to note, DSL is often asyncronous, meaning that your download speed is different than your upload. With fiber these speeds are often the same up and down.
  3. Concider leaving the Server-ip blank in your server config file, I've never had to touch it. And I second Donut's motion of checking the pc running the servers firewall. What OS are you running the server on? An alternative to NOIP that doesn't require something running on your PC is FreeDNS, I've personally used it with great results.
  4. Fair point! is there a nice tool to stress test the drives? My plan was to run crystal disk info to see if there was anything immediately concering, since Amazon doesnt really have the best shipping packaging for HDDs.
  5. Just as an update, I managed to get Samba working on my Ubuntu machine! After bashing my skull against it for 5 hours! The solution was stupid, literally just: sudo ufw allow Samba So, now I'm just waiting for my Ironwolf HDDS and backplane to come in, then I get to experience the joy that is figuring out how to do software raid in Ubuntu and install PLEX. Could you please expand on this? If this doesn't end up working out in the longer run, chances are I'l try freenas. My only dissapointment is that segregating Plex access requires a subscription.
  6. Mentlegen, After some consideration I would like to upgrade my current Linux server to serve a couple additional purposes. Currently it's running modded MC servers but I would like for it to have NAS functionality as well as Plex, mostly plain file storage and acting as a media server. So, my plan is to purchase a backplane and shove it full of ironwolf HDDs, and preferably have the drive array in RAID10. The questions I have are as follows; 1. Would using a Hypervisor benefit me in this use case? I would like to learn to configure Hypervisors as career advancements might make it useful. 2. I would like to do hardware RAID, but I'm unsure about the benefits and downsides to doing so, from what I remember, hardware RAID does not play well with hypervisors and software raid is preferable in this use case. As stated, I'm planning on using raid10 to alleviate single drive failures, but if the raid controller dies, can software rebuild/compile/understand the drives without the controller? or do I need to purchase an identical raid controller? 3. So, what NAS software / OS do you recommend? Can a single Linux OS do both NAS and Plex handling? or should I go for something like FreeNAS and Bundle the MC and Plex servers in a virtualized OS? 4. Can I segregate Plex folders? So one Plex library is available to anyone on my network, and another is only accessible to me? 5. Any benefit to buying Ironwolf HDDs over something more standard in this use case? Current server specs are as follows; Hodgepodge of SSDs and 1 1tb HDD. I5 3470, 16GB DDR3 non-ECC memory. Any help and tips would be greatly appreciated, I've done a fair bit of googling but having some feedback before I spend 500 dollars on drives and controllers would be fantastic.
  7. Fair point, I've just experienced disgruntled friends with 100 Mbps lines complaining about steam "downloading at 12 Mbps!". So I just default to "multiply it by 8". Also a fair point, I have the same type of DSL line, and very often, around 10/1 down/up is the maximum the DSL provider can deliver, and just as often the DSL provider is the only option. Personally my DSL provider is no longer maintaining copper lines as of May, but my street gets gigabit fiber come this fall. Might be a couple rocky months. ?
  8. OP said he had DSL, i'm guessing 100 Mbps is not supplied by the ISP since the phonelines and modem would probably explode ?
  9. ISP's often provide their speeds in "Megabits per second", since "10 Mbps" is a larger number than "1,25 MBps", and people don't notice the small B in Mbps. So heres a small writeup to understand your internet speeds better. Mbps = Megabit per second. MBps = MegaByte per second. Note the large B and Byte vs Bit. 1 Byte is 8 Bits, so the conversion from one to the other is pretty simple. Divide Mbps by 8 to get MBps, and mutiply MBps by 8 to get Mbps. Megabyte is useful cause we often use Bytes to measure filesizes. So lets say youre downloading a 20 gigabyte file at 1.25 MegaByte per second; 20*1000 = 20,000 (turn gigabyte into megabyte) 20,000 / 1,25 = 16,000 (divide size by download speed, giving you seconds) 16,000 / 60 / 60 = 4,4 (divide seconds by 60 to give minutes and 60 again to give hours.) so 20 gigs should take aprox 4-5 hours at 1.25 MBps Also, you used MiB/s, which is Mebibyte per second, which is just the IEC name for megabyte. Just as a last note. Theres decimal and binary multiples of Bytes, so a megabyte can either be 1000^2 Bytes in Decimal, or 1024^2 Bytes in Binary, but MiB is always used for the Binary value. In a realistic case this is such a small difference that it's often overlooked.
×