Computing desk
< October 29	<< Sep \| October \| Nov >>	October 31 >

Welcome to the Wikipedia Computing Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

October 30

Dell rear panel USB

Anyone heard of rear panel USB sockets going intermittent on Dell 2400s? Im having all sorts of trouble with these. --31.55.64.250 (talk) 01:46, 30 October 2015 (UTC)[reply]

I don't know anything specific about Dells, but you might open up the case and look at how the wires connect to the USB ports on the inside. Specifically, if they stripped back the insulation too far, and the wires are touching each other, that could cause bad things to happen. Or, if the wires (with insulation) are sharply bent, that may mean the wire inside has broken. (If both ends are still in contact, that might cause intermittent outages only.) Also, if the connector isn't properly attached to the computer case, it could wiggle too much when you plug and unplug USBs, causing stress on the wires.

Another issue to look for is "USB sag". That happens when the USB drive is loose in the socket and it's weight makes it move out of contact. The cure here is to use a USB extension cord, which allows you to rest the USB drive on the table and support it's weight there. StuRat (talk) 03:29, 30 October 2015 (UTC)[reply]

The sag, combined with a poor connection mentioned in the previous comment, can cause the soldering connection to the motherboard to come loose. When that happens, it needs to be resoldered. However, poor designs like that will just break again (and again). 209.149.114.132 (talk) 13:55, 30 October 2015 (UTC)[reply]

I've seen USB sockets (and many other kinds of frequently-used socket such as audio and power connectors) fail on all sorts of devices - I'd be surprised if Dell had some kind of magical way of making them so they didn't fail - but I'd also be surprised if Dell were any worse than other manufacturers.

Per User:209.149.114.132, it's true that repairing these things is tough. A less pretty way to repair them in such a way that they don't fail again is to remove the bad socket from the circuit board - then take a USB extension cable, cut it in half and solder the wires leading to the "socket" end to the pads on the circuit board. Use a cable-tie to attach the cable to something solid - but leave the socket flapping loose. The flexibility of the wire removes the issue of force applied to the socket being transmitted to the circuit-board tracks through the solder connections. It looks terrible though!

SteveBaker (talk) 14:16, 30 October 2015 (UTC)[reply]

If I was tackling this: I would cheat first, to discount an intermittent dry soldered joint. Use (say) a Freezer spray. The cans are not that expensive when one factors in the time its saves. Plus, there is plenty left in the can for many more uses in the future. Just aim the tube at each joint in turn (with the computer running – as it is electrically safe). If one loses/gains connectivity at one particular joint – then one has located the fault. I am never with out one. Don't use it frequently but when I do, I am thankful for it.--Aspro (talk) 15:15, 30 October 2015 (UTC)[reply]

You can get a used Dell Dimension 2400 for $37.97 plus $26.99 (http://www.amazon.com/Dell-mtc2-Dimension-2400/dp/B00C69DA00/) or you can move up to a Dell OptiPlex 780 (a far more capable PC, yet still in that "cheap used dell that used to sit in an office somewhere" pricing sweet spot) for around $100. (http://www.amazon.com/s/ref=nb_sb_noss?url=search-alias%3Daps&field-keywords=Dell+OptiPlex+780). I have one that I upgraded with a 3.00GHz Intel Core 2 Quad Q9650, 16GB of RAM and a SDD boot drive, and it runs Linux and Windows 10 like a bat out of hell. --Guy Macon (talk) 18:46, 30 October 2015 (UTC)[reply]

If you're going that route and have some mild technical ability, personally i would recommend something with a G43 or G45 (perhaps G41 if you're not worried about the limitations) over the Q45. It's unlikely you need the Vpro, and Qxx chipsets are annoying due to their lack of support of Xeon 5xxx series CPUs. 3xxx series CPUs are more limited and generally more expensive. Upgrading to a Core 2 Quad seems unnecessarily expensive when you can probably spend about half the price for one and get a Xeon which is as good if not better along with the necessary LGA 775 to 771 stickers. Cut off the plastic key bits on the CPU socket, update the microcode and you're done. Nil Einne (talk) 13:35, 1 November 2015 (UTC)[reply]

Before I would invest in a USB extension card in such kind of old PC, I would upgrade for little more money than the card and renew the M/B. A SoC M/B comes with CPU included in the board an price. Just ad a the new RAM. Every thing else inside the Dell 2400 should fit in screw. You might use a new Power supply due aging the cpactiors inside and faster hard disk drive, which might become more reliable within the some months of usage. The SoCs save much energy. CPU included in the chipset is up to 10 watts or less. Check if a FlexATX M/B with AMD CPU and RAM or such SoC is avail for the better price due different memory types and pricing. --Hans Haase (有问题吗) 20:30, 30 October 2015 (UTC)[reply]

"Invest" ? I got my USB extension cords for $1 each at the dollar store. StuRat (talk) 21:26, 30 October 2015 (UTC)[reply]

Wow. You really got ripped off there. I got mine for 99 ¢. and a free voucher promising me 10% off my next purchase. ;-) --Aspro (talk) 22:25, 30 October 2015 (UTC)[reply]

The USB extension cord changes the whole situation on mechanical damages but not on electrical if the 5 volts are affected and the port could not be changed. Onboard jacks are repairable but not easy to be soldered without further damage. --Hans Haase (有问题吗) 22:33, 30 October 2015 (UTC)[reply]

Are we talking cards or cords? --Guy Macon (talk) 22:28, 30 October 2015 (UTC)[reply]

At this stage I don’t think it really matters. The OP has a fault on his PC and it may be down to many things. To diagnose such a problem one thinks of Zebra. In other words, look at the part that fail most frequently. Discount the horses first. --Aspro (talk) 22:47, 30 October 2015 (UTC)[reply]

Not Windows 10

I have Windows 7 on my PC. Is there something I can do (a setting or a registry or even a program???) to tell Microsoft NOT to even try to install windows 10? -- SGBailey (talk) 16:53, 30 October 2015 (UTC)[reply]

Does this help? http://www.windowscentral.com/how-remove-get-windows-10-app-pc --Guy Macon (talk) 18:50, 30 October 2015 (UTC)[reply]

Don't think that's going to work. It is the last day of the month tommorow so I am too busy just yet chopping wood for my end of the month bath. Just give me a few hours though.--Aspro (talk) 19:38, 30 October 2015 (UTC)[reply]

If what you have is a desktop computer, the simplest way to tell Microsoft not to try to install Windows 10 is don't buy the install disk. At this point, you have to pay to upgrade to Windows 10 on a desktop. Robert McClenon (talk) 19:00, 30 October 2015 (UTC)[reply]

? --Aspro (talk) 19:03, 30 October 2015 (UTC)[reply]

SGB is presumably referring to the annoying nag that keeps popping up if you don't install Windows 10. Oddly I was able to remove update KB3035583 from my laptop, but on my desktop machine it keeps reappearing.--Shantavira|^{feed me} 19:20, 30 October 2015 (UTC)[reply]

It's still offering to upgrade for free on mine. I believe they said the free offer will last a year. StuRat (talk) 19:22, 30 October 2015 (UTC)[reply]

@ User:SGBailey Don't act on this until other editors has contributed because Microsoft is up to its shenanigans again and they are masters of misdirection. It is just not Windows update KB3035583 that one needs to consider. See: [1] & [2] noting C:\Windows\System32\GWX. Also, check that Widows 10 has not downloaded in the background and now taking up megagoogles of memory on your system. That needs to be deleted too.--Aspro (talk) 19:36, 30 October 2015 (UTC)[reply]

"It is just not Windows update KB3035583 that one needs to consider" ???? - both links you provided say that all you actually have to do is to remove KB3035583 and make sure it isn't reinstalled.

The first one says they didn't want to do that because it would require a reboot and they wanted to pretend they had the update to Microsoft and also they weren't confident they could stop KB3035583 being re-installed.

The second one is asking about how to block the files installed by the update (the guide the person is using says to install the update!) by other programs; but then other people point out the person could just uninstall the update, which is possibly what the person asking did, it isn't totally clear. (In the second link, someone briefly mentioned the telemetry updates but while there was a big fuss about these, they are unrelated to the OP's problem, unrelated to the GWX directory, and the links you provided aren't useful about them.)

If you're like Shantavira and can't convince KB3035583 to stop reinstalling, then perhaps you have a problem. Personally I think it will generally be possible (exceptions would be where you don't have proper control over the computer, like a work computer), but you could solve it via the other means suggested in those links if you think that's better. Or if you want to install KB3035583 for some reason but disable it (like the person in the first link), well that's up to you.

But the idea that it isn't only KB3035583, isn't supported by the very sources you provided both of which agree all you need to do is make sure KB3035583 is properly uninstalled and not reinstalled. May be it's helpful to read what you're linking to before spreading FUD?

Nil Einne (talk) 13:21, 1 November 2015 (UTC)[reply]

Thx. -- SGBailey (talk) 20:26, 30 October 2015 (UTC)[reply]

xdebug eclipse and centos

Hi there,
I'm trying to debug with eclipse and xdebug a php script.
I've installed xdebug correctly, because it does appear in phpinfo() as follows:

xdebug

xdebug support	enabled
Version	2.2.7
IDE Key	ECLIPSE_DBGP

Supported protocols	Revision
DBGp - Common DeBuGger Protocol	$Revision: 1.145 $

Directive	Local Value	Master Value
xdebug.auto_trace	Off	Off
xdebug.cli_color	0	0
xdebug.collect_assignments	Off	Off
xdebug.collect_includes	On	On
xdebug.collect_params	0	0
xdebug.collect_return	Off	Off
xdebug.collect_vars	Off	Off
xdebug.coverage_enable	On	On
xdebug.default_enable	On	On
xdebug.dump.COOKIE	no value	no value
xdebug.dump.ENV	no value	no value
xdebug.dump.FILES	no value	no value
xdebug.dump.GET	no value	no value
xdebug.dump.POST	no value	no value
xdebug.dump.REQUEST	no value	no value
xdebug.dump.SERVER	no value	no value
xdebug.dump.SESSION	no value	no value
xdebug.dump_globals	On	On
xdebug.dump_once	On	On
xdebug.dump_undefined	Off	Off
xdebug.extended_info	On	On
xdebug.file_link_format	no value	no value
xdebug.idekey	no value	no value
xdebug.max_nesting_level	100	100
xdebug.overload_var_dump	On	On
xdebug.profiler_aggregate	Off	Off
xdebug.profiler_append	Off	Off
xdebug.profiler_enable	Off	Off
xdebug.profiler_enable_trigger	Off	Off
xdebug.profiler_output_dir	/tmp	/tmp
xdebug.profiler_output_name	cachegrind.out.%p	cachegrind.out.%p
xdebug.remote_autostart	On	On
xdebug.remote_connect_back	On	On
xdebug.remote_cookie_expire_time	3600	3600
xdebug.remote_enable	On	On
xdebug.remote_handler	dbgp	dbgp
xdebug.remote_host	77.127.241.252	77.127.241.252
xdebug.remote_log	/log.txt	/log.txt
xdebug.remote_mode	req	req
xdebug.remote_port	10002	10002
xdebug.scream	Off	Off
xdebug.show_exception_trace	Off	Off
xdebug.show_local_vars	Off	Off
xdebug.show_mem_delta	Off	Off
xdebug.trace_enable_trigger	Off	Off
xdebug.trace_format	0	0
xdebug.trace_options	0	0
xdebug.trace_output_dir	/tmp	/tmp
xdebug.trace_output_name	trace.%c	trace.%c
xdebug.var_display_max_children	128	128
xdebug.var_display_max_data	512	512
xdebug.var_display_max_depth	3	3

eclipse keeps stucking in 78%, though it does not stop at any breakpoint. can somebody help? Thanks — Preceding unsigned comment added by Exx8 (talk • contribs) 20:29, 30 October 2015 (UTC)[reply]

Fixed formatting. --NorwegianBlue^talk 11:55, 31 October 2015 (UTC)[reply]

my talk page

Please see my message on my talk page and reply there. --74.130.133.1 (talk) 22:13, 30 October 2015 (UTC)[reply]

To be completely clear about this - we don't do that here at the reference desk. Questions are asked out in the open here - and answers are equally public. Part of the point of this is that people can search our archives and find answers that way. Doing it here has an advantage for you too - people reading this desk will often spot errors and omissions in answers and come here to fix them...you get vastly better answers if you allow "the wisdom of the crowds" to be involved.

So if you have a question - just ask it right here - better for you, better for us, better for our future readers. SteveBaker (talk) 22:51, 30 October 2015 (UTC)[reply]

We also don't answer messages that call for predictions or opinions. --70.49.170.168 (talk) 04:07, 31 October 2015 (UTC)[reply]

... and you keep deleting content from your talk page, so there's not much point in answering there. Dbfirs 12:47, 31 October 2015 (UTC)[reply]