Verizon FIOS Router Actiontec MI424WR-GEN3I Issue - Dropped Connections - Constant Reboots

[g00py]

So, here's my tale of woe: I received a new router (a MI424WR-GEN3I) to replace my old one around August. Almost immediately after setting it up, I noticed that it randomly reboots at least once a day and has been doing so ever since. Half a dozen calls to customer support have been no help whatsoever. I've had my ONT replaced, new routers sent to me, etc. Nothing has remedied the problem.

Today, however, a tech came by and showed me an internal document dated mid-October of last year {edited for privacy} that states Verizon is aware of the problem and the only way to fix it is to disconnect the device on the network that's responsible for dropping the connection. Frankly, I find that an unacceptable solution since, it turns out, this is a known (and apparently widespread) issue with the router itself.

I was told by another tech that Verizon is working on a "patch" that could come at any point in Q1, but I have no way to verify this. Nor do I have any understanding of the release schedule beyond this vague timeframe.

So, my question to the community is this -- has anyone else experienced this problem? 

Oh, and if Verizon is listening, when can we expect a solution? It's been three months since the document referenced above was circulated internally and five months since I started noticing this problem. I think we've waited long enough since this issue essentially makes Verizon FIOS unusable to power users like myself.

[Morningmist]

I have been having same issues since receiving new router / modem from Verizon. Rebooting once a week. The problem seemed to start when we introduced a "smart tv" into the mix. But this could be a coincidence since we received router a month or so before hand. It appears that buying a new router / modem from best buy will not solve issue. Is this correct?

[missy__m]

I think I'm in the same boat. Not quite as often as 30 seconds, but more like 30 minutes. I will be calling Verizon yet again tonight 😞

Drives me nuts. My iMac has the most issues, it rarely connects for more than a minute and when it does my internet speeds are a good 15Mbs slower than they should be, but my Apple TV, iPad and Macbook all disconnect frequently. The entire network disappears though, sounds slightly different to what most of you are dealing with ...

😞 I thought I'd escaped from Time Warner internet woe, but this is surprisingly even worse...

[g00py]

Won't be able to call support until tomorrow, but in the meantime I've assigned static IPs to all my devices and I'm considering port forwarding too. Not sure if it'll make a difference, but I've tried everything else. Nothing to lose at this point.

missy_m: Just a piece of advice -- escalate to tier 3 support if you haven't already. 

[TheHangingJudge]

Same problems here with my gen3 rev i.  Frequency is about 3-4 days between reboots so not horrible.  2 calls to support.  1st changed router out.  Second one changed ethernet cable.  I am ethernet to the ONT.  I do have 2 cisco STBs and a cisco DVR.  This has been happening for 3-4 months.  It must be due in part or whole to the adaptec firmware 40.21.10.2.

2 - Cisco chs335hdc Release number 1.9.5 Build number 17.17

1 - Cisco chs435HDC Release number 1.9.5 Build number 17.17

If more people post if they are having this problem VZ will take note.

[sangs1]

---

If more people post if they are having this problem VZ will take note.

---

Between here and the DSLReports forums plenty of people have been posting about it. Do a search and you'll find some of the posts date back more than 18 months. So far, VZ hasn't really taken note. 

[jphilly37]

I am having the exact same issue, my reboots are random.  I called tech support today and spoke to two people.  The first RIGHT AWAY told me about the issue with the Actiontec Router and the Cisco boxes.  I then complained and asked to talk to a supervisor.  Basically told the supervisor that they should be reimbursing their customers for this un-reliable service.  He told me that they issue has been escalated to their engineers and that they are working with the hardware manf.   Do I believe this? Maybe... do I think it will get resolved soon?  No way!

Since I'm a brand spanking new customer I have 30 days to cancel.  I told them they  have a week to show me their reliable network.. .otherwise I'm going back to comcast.

Stinks, cause I actually like FIOS, but I CANNOT have my internet dropping on me when I'm dealing with work issues.

I have no idea how their isn't a class action against them for false advertising.  Their network is not reliable and your uptime is no where near 99%... get a lawyer guys.  This is customer service at it's worse.

[sangs1]

This whole Cisco excuse doesn't fly though, unless Cisco STBs & DVRs are being rebranded as Motorola. Because many of us experiencing this have NO Cisco equipment whatsoever.

[jphilly37]

@sangs, I hate to agree with you, but I do.. and I think that means I should just go ahead and cancel.  That sticks...  I have no idea how this hasn't hit the media yet.  This is a huge story and the fact that they know something is wrong and have no answers??  Wow.

[sangs1]

I understand the frustration jphilly, believe me, but ease back a bit.

1) Until the last 1-1/2 months my FiOS service was pretty much perfect. No dropouts, no outages, no reboots, nothing;
2) They're trying to figure this out. They've jumped through hoops for me attempting to sort this out - and they've been doing it quickly;
3) This is not a media story. It doesn't impact a big group of people for one and it's not like this has been happening consistently since FiOS came onto the scene about 10 years ago.

It's aggravating, frustrating and maddening - and we have every right to complain about it - but we also have to hope a fix is coming. If not, people will switch and then the bottom line starts being impacted.

[jphilly37]

@Sangs  Yep, I hear you and perhaps I am a bit annoyed.  I guess it all doesn't make much sense to me.  From what I've read we all have ONT boxes in our garages/basements with a disabled ethernet port.  Since they don't have an ETA.. Enable those ports offer up some free wifi routers and have people up.  Since the issue seems to be on the Coax side, this seems like a customer friendly solution.  I guess I'm too close to customer service since I'm in IT and this would just never fly with my customers.  All in all I get it.  Last post, I promise!

[TheHangingJudge]

Getting the ONT ethernet port enabled only takes a phone call to support.  Switching to the coax port was actually one of the troubleshooting steps that I tried with VZ support.  It didn't work so I am back.  

By the way, I thought I would try something constructive to find the problem. Here is my first tidbit to chew on after about 5 minutes of comparing the Actiontec firmware source code from 40.20.4.2 to 40.21.10.2.

This line of code exists in the newer firmware.  I was told that VZ tried this and removed it as it didn't test well.  While this isn't definitive, it could explain why the router is rebooting itself, if it is responding to a self heal function.

/* Actiontec Verizon self heal function defines */
{ "ACTION_TEC_VERIZON_SHEAL", NULL, OPT_H | OPT_HC | OPT_MK },

Just something to chew on!  Enjoy and please discuss.  I will report back on more findings.

[g00py]

I've been told the self-healing feature is the cause and then, two weeks later, that it wasn't because "it doesn't work that way." I've also received conflicting information about UPnP and the "half-closed socket" issue. At one point I was told that a device on my network was the culprit. And for weeks Verizon insisted other components -- ONT, coax, splitters, etc. -- were to blame. (They've since been replaced.) And now, it seems Cisco boxes are the issue. Anybody care to guess where the roulette wheel will land next? At least -- to sang's point -- customer service has been trying. Still, that doesn't make the wildly inconsistent information I've received any less frustrating.

[TheHangingJudge]

Well I am only trying to shed some light on the problem from one angle.  The actiontec firmware changes from 40.20.4.2 to the current version 40.21.10.2.  I don't even recall when or if my router was running the ealier firmware, but that was the prior version on actiontec's website.

After comparing all the source files from the two version, the major changes are summarized below.  There are many new files included in the new firmware source code that would take more time to review.  Most of it seems to be related to IPv6.

In new firmware
Verizon Self Heal
IPv6

Removed from new firmware
IGMP is handled differently (Hard to say how.)

I realize this probably raises more questions than it answers, but It at least provides more information.  For me, I need to look into the source code more to find some more information about the Self Heal function.  It seems like the best starting point.

[g00py]

There's definitely value in what you're doing and, for what it's worth, self-healing could very well be the issue. Who knows? It seems a more likely cause than anything else Verizon has said to date.

[TheHangingJudge]

---

@g00py wrote:

There's definitely value in what you're doing and, for what it's worth, self-healing could very well be the issue. 

---

Thanks...I seem to have reached a dead end.  There doesn't appear to be any more information about the Self Healing function in the firmware (that I can find).  

Only other thing I found was a phase 2 implementation of tr-143.  Tr-143 is essentially performance testing for broadband service providers to gauge the quality of service based on performance norms.  I don't know what VZ's 2nd phase of that may be.  

While I can't put my finger on the exact cause, it seems very likely to be caused by the changes I mentioned earlier.  It could be something to do with IPv6 implementation, but more likely the Self Heal function.  

Hopefully someone will respond with a solution from VZ!

[g00py]

Interestingly, self-healing was the first thing Verizon said when I started making calls back in August, but the story's gotten way murkier since then.

[TheHangingJudge]

---

@sangs wrote:

---

If more people post if they are having this problem VZ will take note.

---

Between here and the DSLReports forums plenty of people have been posting about it. Do a search and you'll find some of the posts date back more than 18 months. So far, VZ hasn't really taken note. 

---

Yeah in my mind I thought MAY instead of will.  Just typed the wishful thinking.  

[g00py]

As usual, Verizon's understanding of the problem is only partially correct since many of us don't even have Cisco boxes.

[Josi]

Oh yes I have the same problem. This started to happen the day I updgraded to Quantum 50/25 and got a new router. It never happened with the older  router which I used when I had 25/15. I only got the new router because it defnitely produces an improved wireless signal. I have an appointment with  them on the 18th.  We will see I will keep this forum posted

[chaybee1]

I believe you will run into a problem if you switch back to the old router with the quantum speed.  It is my understanding that only the rev "I" can handle it.  I came across this first hand as I did not read the fine print when I upgraded to Quantum.  I was streaming a baseball game on MLB.TV and it kept buffering and reverting to Stanard Def.  After looking into this, I was informed that the old router "E" i think? cannot handle HD and 50/25 speeds.  Of course though,  once I upgraded, the random reboots began.   

[g00py]

True...older routers are definitely limited. But you could connect the newer router to the old one and potentially avoid this problem. That is, run coax directly to the older router, then connect the newer actiontec to the first via ethernet and stream off of that. I would have done this myself a long time ago, but I'm using my older router as a moca bridge in an entirely different room and have four different peripherals hardwired to it.