Drive completely destroys my network

So here is my setup:

Frontier provided Westell router: 4 ethernet ports, WiFi shut off. Router is providing IP and DNS for the entire network.

connected to the router:

iMac (dynamic IP)

iomega NAS (static IP)

WD NAS (static IP)

Airport Extreme (static IP for Airport and anything that connects to it by the WiFi network is dynamic IP)

When the WD NAS is on the network performance goes completely to **bleep**. I don’t just mean speeds. I mean devices can no longer connect to the network, devices drop from the network, internet connections stop responding.

It’s so bad that even opening the router configuration page and trying to look at settings is painfully slow. As soon as I shut off the WD NAS it’s fast as you would expect a direct connected device to be.

All I have to do to fix any of the issues is simply turn the WD NAS off. Then everything works fine. I can’t believe this one device is causing so much trouble on my entire network. It makes it impossible to get anything done and causes problems for everyone in my family. This is not what I expected, nor what is acceptable, for a $500 network drive.

On top of it all I submitted a service report last week and got back the auto-generated response telling me that I would hear from WD in about 24 hours. Now almost a week later I’m still waiting. Really? Why do I feel like I just got ripped off? 

Hello RussW.

I have the same problem.

It seems that some routers support WD NAS network communications problem, some other not like you and me.

I’m also waiting the next firmware update and hope that it will solve this.

You can try a firmware update on your router (if not done).

Also, in my configuration, I only disconnect the NAS from the network and reconnect it when I want to use it.

After some days, I think some bad communcations goes to timeout and I can use my network, but have no access from the web.

RussW:   

Are ALL of the “client” systems WiFi connected?   

If not, do these same problems happen to the WIRED clients?

What I’m getting at is, is there any possibility that one of the WiFi devices is doing something with the NAS such as transferring files or doing automated backups, and that it might be hogging all the WiFi bandwidth?

(Remember that WiFi is “shared” among all the computers on the same channel, whether they’re actually on your network or not.)

I can’t imagine the MBLD doing ANYTHING that would create that kind of havoc other than what I describe above.

But if you’re curious and have the wherewithal to try it, you could run Wireshark on one of the wired PCs and get a capture to see just what’s going on your network that would cause these issues.

Tony,

I totally understand what you mean when you can’t imagine the MBLD causing all these issues, I felt the same way until I did some more trouble shooting of my network. 

Last night my Apple TV box wouldn’t connect to my network. It wasn’t that it just couldn’t access the internet, it wouldn’t connect to my WiFi network at all. After simply shutting down the MBLD no problems. The Apple TV connected on the first try and stremed content from both my iMac and the net.

Yesterday I even unhooked everything from my router except the iMac and the MBLD. With the MBLD turned off there were no issues. I could access the internet and the configuration interface for my router responded very quickly. Then I turned the MBLD on. After it was booted up an running the configuration interface fo the router was painfully slow to refresh and my internet was also. No WiFi involved, all wired connections. 

There are no devices on my network that use the MBLD for any type of automated back up or media streaming source. As a matter of fact I have all forms of media streaming and remote access on the MBLD shut off. All I use this drive for is my picture library. There is nothing on the drive but image files. Even with this being the case the green activity light blinks constantly.

I also started a full diagnostic test on the MBLD last night at around 7pm. As of right now, 7:20am it’s only 90% done. Should it really take over 12 hours to run an internal check on the MBLD?

I’ll look into Wireshark and see what I can find.

Tell us if you also see some broadcast DHCP request (what I see when I analyse my network with Wireshark).

Is WD interested by log of data (I can plug a simple hub in my network and so check everything on it)

And yes, analyse the disk takes a lot of time.

Makes me wonder if the MBL is just sending some sort of “Packet of Death”  (Loose term, by the way…) that’s affecting your router.

If you have a buddy that has a different brand of router, see if he/she’d be willing to let you borrow for a few hours to see, or see if you can take your MBL to their home and try it there.

The MBL *does* do some things that, if the router isn’t prepared for it, may cause those issues.  Thinks like LLDP or LLTD (two network discovery protocols), multicast (mDNS) or such can wreak havoc on devices that aren’t prepared for them.

A router should be able to handle any packets. None of the things the MBL Duo does is non-standard. If a router cannot handle it, then it is likely a bug in the router code.

WDTony - 

If it’s a router issue then 2 different routers and 2 different Airport Extremes all have the same bug.  Sorry, not buying that. There is an issue with the drive.

I made a mistake in my orginal post. My current router from my ISP is an Actiontec MI424-WR. It’s running the current firmware. This issue has been present with this router, the Westell 9100EM I had before and now 2 different Airport Extremes (I bougt a new one yesterday just to rule it out).

At this point I’m wondering if I can hook up another switch and use it to segregate the router from the rest of my network. I really just need the router to work with my iMac and only as a storage/backup for photos. Could I hook up the iMac and MBLD to the switch and configure the switch to only let traffic from the iMac out to the rest of the newtork? This seems nuts to have to goto this extreme just to use this drive but if I can’t solve this issue then I pretty much have a $500 paperweight and I’m going to be really pissed.

On a side note. I’m now at 16 hours to try and run a full diagnostic on the MBLD. It’s been sitting at 90% done for over 4 hours. This is just plain stupid. The drive is only 5% full.

That’s interesting.   I’m guessing you have Verizon FiOS, because those are both FiOS specific routers.

I’m on FiOS, too, and used to have a Westel (a long time ago) and now have the same MI424-WR you have.

No issues at all on mine, and I’ve had as many as four MBLs running simultaneously; I have three online right now.

The thing with BOTH of those routers are:

  1-  The internal switch is 100meg only.

  2-  The wireless is B/G only.

So they’re both slower-than-dirt – even when using them on the local network only.

But here’s my difference… Because those routers are so slow, I don’t use the router’s internal switch or WiFi.   I have a gig-E switch that connects everything else in my house, and the MI424-WR is connected to that switch.  I also use separate Cisco E3000 A/B/G/N WiFi boxes, all of which are Gigabit Ethernet attached.

So, other than what I describe above (about some incompatible packets, which is ruled out because you and I have the same routers), what I said before still suspect… 

SOMETHING must be consuming your resources on your network … apparently something that you may not be aware of.

There’s nothing the MBL is doing that affects my Actiontec MI-424WRs (running current firmware:  4.0.16.1.56.0.10.14.4 running on a Revision E platform)  

Tony - 

How is it only consuming network resources when the MBLD is connected? When I shut the MBLD off everything works perfectly.

If this diagnostic scan ever finishes, going on 17 hours now, I’m going to put the switch in and see if it makes any difference.

RussW wrote:

Tony - 

 

How is it only consuming network resources when the MBLD is connected?

The presumption (yes, presumption, not fact…) is that the target being used IS the MBLD.

There’s actually a farily easy test to look at.

Log into the MBLD via SSH (see the forum for how to do this)

Execute the command

ifconfig eth0

example:

MyBookLive:~# ifconfig eth0
eth0 Link encap:Ethernet HWaddr 00:90:a9:f6:05:f8
          inet addr:10.0.0.15 Bcast:10.0.0.255 Mask:255.255.255.0
          inet6 addr: fe80::290:a9ff:fef6:5f8/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
          RX packets:62644473 errors:0 dropped:0 overruns:0 frame:0
          TX packets:24974502 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:436205658 (415.9 MiB) TX bytes:1616656110 (1.5 GiB)
          Interrupt:36

 Note the RX PACKETS and TX PACKETS numbers.

Wait 10 seconds, do the command again.

Note the new values, and subtract the smaller from the larger.

That’s how much network was used during 10 seconds.

Let us know what you find.

If the number is substantial, you can then execute this command:

lsof | grep EST

(Note, that’s CASE SENSITIVE.)

That will list all the active connections using the MBL and what those connections are using.

OK. So I got sick of waiting for the diagnostic to finish. I think 17 hours was a bit too long. I cancelled it.

Next I hooked up an ASUS Gigabit, 5 port switch to my FIOS box. Then I plugged in the network connection for my iMac. All good. I have internet, everything is quick. No problems. Then I hook up my Iomega NAS, boot it up, check it’s status, no problems. I still have net access, it’s still quick, all good. Then I connet and start up my MBLD. Wait for it to boot up, check it’s status, all good.

Not even five minutes later…

Can’t connect to the net on my iMac. Says I have no network connection. I have made NO CHANGES since starting up the MBLD.

So I shut down the MBLD…

Instantly the iMac is back on the net. Fast, any site I goto, no problems.

I"ve had it with this thing. I’m making a call to support and if they can’t figure it out I guess it goes in the garbage and I’m out $500.

Oh and by the way, I still haven’t recieved a response form the support request I sent in a week ago. I’m not getting my hopes up for phone support.

Tony, I agree with you that router should support everything, and bad packets.

But some of them don’t support, and I suggest that WD analyse what is wrong with network and stop sending wrong packets on it.

I think you have to take care about customers who can’t change of router and who want to use WD products with all functionalities work OK.

I’m in contact with router company trying to understand what is wrong (We supposed something about DNS, and I work in this company as a computer engineer, so it is more easy for me to have a good contacts with development guys).

But right now my network doesn’t work well and I can not acces to my router from the internet.

I think Russw has same type of problems.

The only thing I would like is you say us ok, something is wrong, and we try to have positive action that may help you to have our product work well, than saying it is a problem in the router, we don’t care about it.

I think problems are in some routers (and you do some tests OK with good routers) AND in the WD product

My result about the command you suggest

MyBookLiveDuo:~# ifconfig eth0
eth0      Link encap:Ethernet  HWaddr 00:90:a9:b7:1f:1b
          inet addr:192.168.1.18  Bcast:192.168.1.255  Mask:255.255.255.0
          inet6 addr: fe80::290:a9ff:feb7:1f1b/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:10267500 errors:0 dropped:0 overruns:0 frame:0
          TX packets:5028873 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:778420925 (742.3 MiB)  TX bytes:1899633186 (1.7 GiB)
          Interrupt:36

MyBookLiveDuo:~# ifconfig eth0
eth0      Link encap:Ethernet  HWaddr 00:90:a9:b7:1f:1b
          inet addr:192.168.1.18  Bcast:192.168.1.255  Mask:255.255.255.0
          inet6 addr: fe80::290:a9ff:feb7:1f1b/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:10267581 errors:0 dropped:0 overruns:0 frame:0
          TX packets:5028924 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:778430241 (742.3 MiB)  TX bytes:1899654939 (1.7 GiB)
          Interrupt:36

How do you log to the drive?

I enabled ssh and can get terminal open on my mac but don’t know how to go any further.

So I have now been waiting for an hour on hold for “support”. I wonder if I’ll ever buy another WD product again. I should have just spent the extra money and got a Drobo.

RussW wrote:

How do you log to the drive?

 

I enabled ssh and can get terminal open on my mac but don’t know how to go any further.

    • *I think on a Mac you can just use the command “ssh x.x.x.x” where x.x.x.x is the IP of your MBL. When prompted, enter the userid and PW the MBL told you to use when you enabled SSH.

Oh, by the way… Just thought of something else… Everything on your net has dynamic IPs except your MBL, router, and airport. This may sound stupid, but is there any chance you accidentally used the same ip on your MBL that you already used on another device? Or did you perhaps assign the static in the same range as the Actiontec DHCP range? My actiontec only assigns DHCP in the range of 100-254, so all of my statics are in the range of 2-99.

so when i type “ssh x.x.x.x” at the command prompt it asks me for a password. i enter the password i got when turning ssh on for the MBL and it says it’s the wrong password. it doesn’t give me any request for the ssh user id.

Default login and password are root and welc0me (with a zero and not O character)