Superhub Annoying quirk - SHUB 2 - Have to sign in twice?

Risco · 16-10-2013, 23:06

Has anyone noticed this on Safari with Mac? Sometimes it will sign in first time, and sometimes I have to enter my password twice!

Sephiroth · 17-10-2013, 11:21

That's another well documented annoyance. The VM SH2 Trial Forum had pages of threads on the subject. The phenomenon has abated (not gone away) with v26 firmware.

I was told by one of the actual firmware bods that it was down to the implementation of stringent VM's security requirements and was a timing decision in the firmware.

As I said, it's abated due to firmware changes. But I still get it occasionally in other places, for example being sent back to the login screen if I try to change router/modem mode. I can try several times (sometimes) before the change I want has been actioned.

17-10-2013, 12:31

I get that - annoying indeed but not as annoying as the usage meter resetting every 4GB. Which I also get.

General Maximus · 17-10-2013, 18:47

Quote:

Originally Posted by Sephiroth

The VM SH2 Trial Forum had pages of threads on the subject.

i did apply to be on the trial when news first broke and I am flipping glad I wasn't accepted now. It would do my nut in if I reported genuine visible problems on a "trial forum" and they did sweet fa to fix it. If you was the dude in charge of the shub2 rollout I would be sitting down with everyone and saying "right guys, I don't want us to make the same mistakes as we did with shub1 so I want a list of all the bugs people are reporting with shub2 and we aren't going to release it to the general customer base until we have got them all fixed and a new firmware update tested and released".

How hard can it be?

qasdfdsaq · 17-10-2013, 19:11

Since it's impossible to ever fix *all* bugs in anything but the simplest of codebases, the answer is... pretty hard. Businesses also do not have infinite time or money so it's logical to prioritize the biggest and most important issues and leave or ignore the smaller/less critical ones.

TBH I'm still impressed by how much improved the wireless is - in comparison, a minor login glitch while doing something that most people rarely if ever do, on a lesser used browser, on a lesser used operating system, is a complete non-issue by comparison.

Plenty of websites that rely on user logins to do anything useful at all still have glitches and issues with certain browsers, including ones more common than Safari.

Sephiroth · 17-10-2013, 19:28

Quote:

Originally Posted by General Maximus

i did apply to be on the trial when news first broke and I am flipping glad I wasn't accepted now. It would do my nut in if I reported genuine visible problems on a "trial forum" and they did sweet fa to fix it. If you was the dude in charge of the shub2 rollout I would be sitting down with everyone and saying "right guys, I don't want us to make the same mistakes as we did with shub1 so I want a list of all the bugs people are reporting with shub2 and we aren't going to release it to the general customer base until we have got them all fixed and a new firmware update tested and released".

How hard can it be?

There was discussion on that too. We said at the time that they would come in for stick with R19. Their focus for R26 (which went public 1 month after the triallists got it) was to deal with the latency variations and 12 hour T4s.

For my part, my SH2 was wholly stable within less than one month of trial commencement (I was in router mode). To debug the T4 events, I was one of the users who received a Netgear specialist at home. But it wasn't stable for all triallists; those on Motorola BSRs had instability until the pre-release candidate for R26 (R23).

Decent wireless as compared with SH1 was a feature right from the start of the trial.

So there's a potted history.

Risco · 17-10-2013, 20:21

Quote:

Originally Posted by Sephiroth

That's another well documented annoyance. The VM SH2 Trial Forum had pages of threads on the subject. The phenomenon has abated (not gone away) with v26 firmware.

I was told by one of the actual firmware bods that it was down to the implementation of stringent VM's security requirements and was a timing decision in the firmware.

As I said, it's abated due to firmware changes. But I still get it occasionally in other places, for example being sent back to the login screen if I try to change router/modem mode. I can try several times (sometimes) before the change I want has been actioned.

Ahh, not just me then! Glad it is a bug!

Quote:

Originally Posted by everyday

I get that - annoying indeed but not as annoying as the usage meter resetting every 4GB. Which I also get.

I get that too!

horseman · 18-10-2013, 08:40

Quote:

Originally Posted by qasdfdsaq

Since it's impossible to ever fix *all* bugs in anything but the simplest of codebases, the answer is... pretty hard. Businesses also do not have infinite time or money so it's logical to prioritize the biggest and most important issues and leave or ignore the smaller/less critical ones…

Good ol Pareto rule! Absolutely - and I was even tentatively "warned off" pursuing some more esoteric ARP/DHCP (timeout/garbage collection) issues further on Trial forum as they understandably didn't represent a "show stopping" issue and wouldn't even be detectable by the majority of users anyway and apparently were at risk starving firmware teams limited resource by investigating further!

progers · 18-10-2013, 09:34

I very rarely get it now, keep your browser cache cleared regularly, seems to help

Sephiroth · 18-10-2013, 09:58

Quote:

Originally Posted by progers

I very rarely get it now, keep your browser cache cleared regularly, seems to help

Yes - that was one of the things alluded to me by the Netgear bod. IIRC, he said that VM had spec'd the firmware to be very tight on sign-on security and that included such things as hackers making use of cached information. In the early days of clearing the browser cache, it still didn't help much. Today, with firmware changes, it may be a vestige effect. It happens so rarely now to me (because I'm now in modem mode) that I'm not bothered.

progers · 18-10-2013, 10:17

What Seph said

qasdfdsaq · 18-10-2013, 13:17

Quote:

Originally Posted by horseman

esoteric ARP/DHCP (timeout/garbage collection) issues further on Trial forum as they understandably didn't represent a "show stopping" issue

Did these relate to the SH1 or SH2? I recall the SH1 having serious problems with any ARPs it wasn't expecting or changing interfaces which made it unusable for me, I wonder if they were the same ones you found.

horseman · 18-10-2013, 16:44

Quote:

Originally Posted by qasdfdsaq

Did these relate to the SH1 or SH2?

Unfortunately it was SH2 related and thus detailed on Trial forum so I can't reproduce it verbatim but if you have access then the link is >
SamKnows connectivity, latency and SH2 in RouterMode
Briefly it related to symptoms of wired (24x7)devices resetting their individual "connection times" due I believe to ARP/IP ageing algorithms configured in SH2 Linux kernel/firmware when no traffic was passing.

Because I was "politely" (and very sweetly) persuaded not to pursue my questions, I simply stopped effort in diagnosing further. However I was postulating a similar theory for some sporadic wireless connectivity(drop-out) issues and resource contention along similar lines.

Sephiroth · 18-10-2013, 17:24

He's good, isn't he?

qasdfdsaq · 18-10-2013, 18:40

Quote:

Originally Posted by horseman

Unfortunately it was SH2 related and thus detailed on Trial forum so I can't reproduce it verbatim but if you have access then the link is >
SamKnows connectivity, latency and SH2 in RouterMode
Briefly it related to symptoms of wired (24x7)devices resetting their individual "connection times" due I believe to ARP/IP ageing algorithms configured in SH2 Linux kernel/firmware when no traffic was passing.

Because I was "politely" (and very sweetly) persuaded not to pursue my questions, I simply stopped effort in diagnosing further. However I was postulating a similar theory for some sporadic wireless connectivity(drop-out) issues and resource contention along similar lines.

Interesting. Though different to my issues then, since on the SH1 the ARP tables would get borked even with live traffic passing. Glad to see they're improving things all round