If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below. |
|
|
|
Thread Tools | Display Modes |
#1
|
|||
|
|||
Surprise!
I was innocently doing some Google searching with the Srware Iron browser
and had just clicked on a Walmart site. While waiting for it to load I looked away to do something else and when I looked back the computer was booting! I vaguely recall a similar reboot many many years ago but not again for many years. Something to be concerned about or just a seldom-seen XP bug? If it matters using Kaspersky virus protection. XP Home with SP3 TIA -- I love a good meal! That's why I don't cook. |
Ads |
#2
|
|||
|
|||
Surprise!
KenK wrote:
I was innocently doing some Google searching with the Srware Iron browser and had just clicked on a Walmart site. While waiting for it to load I looked away to do something else and when I looked back the computer was booting! I vaguely recall a similar reboot many many years ago but not again for many years. Something to be concerned about or just a seldom-seen XP bug? If it matters using Kaspersky virus protection. XP Home with SP3 TIA Does the Kaspersky have a log ? Paul |
#3
|
|||
|
|||
Surprise!
Paul wrote in news
KenK wrote: I was innocently doing some Google searching with the Srware Iron browser and had just clicked on a Walmart site. While waiting for it to load I looked away to do something else and when I looked back the computer was booting! I vaguely recall a similar reboot many many years ago but not again for many years. Something to be concerned about or just a seldom-seen XP bug? If it matters using Kaspersky virus protection. XP Home with SP3 TIA Does the Kaspersky have a log ? Yes, but the three threat logs are all empty. So evidently the boot was not caused by a virus? I don't see any other applicable logs. Paul -- I love a good meal! That's why I don't cook. |
#4
|
|||
|
|||
Surprise!
KenK wrote:
Paul wrote in news KenK wrote: I was innocently doing some Google searching with the Srware Iron browser and had just clicked on a Walmart site. While waiting for it to load I looked away to do something else and when I looked back the computer was booting! I vaguely recall a similar reboot many many years ago but not again for many years. Something to be concerned about or just a seldom-seen XP bug? If it matters using Kaspersky virus protection. XP Home with SP3 TIA Does the Kaspersky have a log ? Yes, but the three threat logs are all empty. So evidently the boot was not caused by a virus? I don't see any other applicable logs. Then you're off to Event Viewer for a look. ******* Does the Walmart site allow third-party advertising ? Like, with a Flash exploit in it ? Paul |
#5
|
|||
|
|||
Surprise!
KenK wrote:
I was innocently doing some Google searching with the Srware Iron browser and had just clicked on a Walmart site. While waiting for it to load I looked away to do something else and when I looked back the computer was booting! I vaguely recall a similar reboot many many years ago but not again for many years. Something to be concerned about or just a seldom-seen XP bug? If it matters using Kaspersky virus protection. XP Home with SP3 Sometimes you may not notice power glitches (outages). They are short to you, maybe so short you won't see the lights flicker, especially if it is daytime and you have no lights turned on, but the outage duration is long enough that the capacitors in the power supply in the computer cannot cover the flicker in power. 300 ms is too long for a PSU but very short to you. Capacitors as well as everything else in a PSU will age and not perform as well over time. Also, brownouts can cause the PSU to shutdown and those can be momentary. Some UPSes will compensate for brownouts (for awhile but not indefinitely) by supplanting battery energy to what is obtained from the wall outlet. Some don't. Do you have a UPS to which your computer is attached? If so, either it is undersized or its battery is too old and needs to be replaced. If not, you're exposed to those power flickers. Most, maybe all, UPSes have software you install to control the power state of the computer and communicate to the UPS. The point of a UPS is to prevent slamming down the computer in a power outage but to shut it down gracefully. The UPS software will keep the computer up but only for so long depending on the remaining capacity and load on that is reported by the UPS. The software will perform a shutdown when there is not enough remaining run-time in the UPS. With an undersized UPS, its software client will perform an almost immediate shutdown of Windows during a power outage. Some software for UPSes will mention when they are correcting for a power outage or brownout but not all do. If the remaining capacity for the UPS is too low, there won't be any uptime left on the UPS so the software has to perform an immediate graceful shutdown in an attempt to preserve data integrity. A right-sized UPS is for data protection. An oversized UPS is for convenience in keeping the gear up longer than needed for data protection; i.e., for you to keep using the computer. Alas, many users buy under-sized UPSes. Also, with age, the battery in the UPS will have less capacity, so one that started oversized may become undersized. I had a 2 kVA UPS that gave me 20 minutes of uptime after a power outage (would've been longer with a smaller load). After 5 years, that was down to around 2 minutes. I replaced the batteries (which constituted 60 pounds of the total 80 pounds of weight for the UPS) and got back up to the full uptime. While some UPSes have an alarm that sounds when the UPS is supplying power (whether for a full outage or to supplement during a brownout), not all do. Even for those with alarms, users may configure the UPS or the software to not sound off during a power event. The alarm on mine was very loud so I put some tape over the speaker to muffle it a bit. If you have a UPS, and if its software has logging and logging was enabled, check that log to take note of power events (outages, brownouts, etc). See if one of those power events coincides with when your computer did a shutdown. |
#6
|
|||
|
|||
Surprise!
VanguardLH wrote:
KenK wrote: I was innocently doing some Google searching with the Srware Iron browser and had just clicked on a Walmart site. While waiting for it to load I looked away to do something else and when I looked back the computer was booting! I vaguely recall a similar reboot many many years ago but not again for many years. Something to be concerned about or just a seldom-seen XP bug? If it matters using Kaspersky virus protection. XP Home with SP3 Sometimes you may not notice power glitches (outages). They are short to you, maybe so short you won't see the lights flicker, especially if it is daytime and you have no lights turned on ATX supplies have "holding time". This is listed in the specs. Just grabbing a PSU at random from Newegg... "Seasonic SSR-1000TD PRIME 1000W" https://www.newegg.com/Product/Produ...9SIA3AR5N20521 "Hold-up Time (20 ms)" That's 20 ms when supplying 1000W of DC output power. You don't normally run them flat-out like that. Normally, the PC is running at a fraction of full load, and you can easily get 10x the holding time in idle desktop situations. So the supply can last for 200ms. Now, when is this important ? Take the situation when you own *two* PCs. The first PC is running. The second PC is plugged into the same power strip. You flip the switch to ON, on the back of the second PC. Instantly, the AC is flattened for two or three AC power cycles. If it wasn't for "holding time" imposed by the capacitor on the ATX supply of the first PC, it would have crashed! But it doesn't crash. The power surge caused by charging the main cap on the second PC, doesn't cause the first PC to reset or anything... Really quite amazing. So yes, a 1 second outage, the kind the power company likes to make when doing things in the switch yard, the PC can drop on one of those. You want a UPS to help you withstand 1 second outages. For tiny load glitches, the 200mS in that ATX capacitor can smooth things over. But for transient power events, as long as the main cap in the ATX supply doesn't run all the way down, the design can handle that. That's why things in your computer room, aren't quite as grim as they could be. I've worked with gear that had abnormal holding time characteristics, and I can tell you, it's hell. Any burp or fart, and there's trouble. The ATX supply has room for a good sized cap on the mains side. C5 and C6 in this schematic, provide "hold-up" on the primary side of the switcher. The 220K resistors next to the caps, are the safety bleeder resistors. http://www.pavouk.org/hw/en_atxps.html Paul |
#7
|
|||
|
|||
Surprise!
Paul wrote:
VanguardLH wrote: KenK wrote: I was innocently doing some Google searching with the Srware Iron browser and had just clicked on a Walmart site. While waiting for it to load I looked away to do something else and when I looked back the computer was booting! I vaguely recall a similar reboot many many years ago but not again for many years. Something to be concerned about or just a seldom-seen XP bug? If it matters using Kaspersky virus protection. XP Home with SP3 Sometimes you may not notice power glitches (outages). They are short to you, maybe so short you won't see the lights flicker, especially if it is daytime and you have no lights turned on https://www.newegg.com/Product/Produ...9SIA3AR5N20521 "Hold-up Time (20 ms)" That's 20 ms when supplying 1000W of DC output power. You don't normally run them flat-out like that. Normally, the PC is running at a fraction of full load, and you can easily get 10x the holding time in idle desktop situations. So the supply can last for 200ms. As I said, about 300 ms. I was being generous. Also the older the PSU, the less holding time it will have for power flutters. That average reaction time of a human is 750 ms, shorter when prepared. A 20 to 300 ms power flutter may not be noticed by a human, especially during daytime hours when lights aren't on or washed out by daylight. |
#8
|
|||
|
|||
Surprise!
VanguardLH wrote:
As I said, about 300 ms. I was being generous. Also the older the PSU, the less holding time it will have for power flutters. That average reaction time of a human is 750 ms, shorter when prepared. A 20 to 300 ms power flutter may not be noticed by a human, especially during daytime hours when lights aren't on or washed out by daylight. I can see the three-cycle glitch caused by a PC being switched on at the back. It's not that hard to catch. I could see it, with an incandescent light bulb in the overhead light fixture. You can't see it with LED lighting, because of the regulated nature of a lot of LED lights. (Only some of the cheaper ones, use a capacitor to limit current flow, and those suck because of their limited tolerance to electrical environment.) I have one PC, where the transient from the power switch is so bad, it causes a reaction from the UPS. The UPS appears to be approaching overload. My UPS doesn't have an LCD panel, as those have more informative displays to tell you what just happened. Anandtech did a recent review (last six months) of a 1000W supply, and when it switched on, the transient tripped a breaker. The claim was, the design was missing the inrush limiter feature on the Active PFC circuit. On 80+ supplies, the Active PFC can perform two useful functions. It brings the power factor back to unity, and during startup, it acts as an "RC" circuit in terms of draw from the mains. And that particular supply under review, the PFC implementation didn't have the inrush feature in it. Normally, cheap supplies (ones without active PFC), they use a NTC resistor to limit startup current. That would be NTCR1 in the upper left. The resistance is zero, when NTCR1 has heated up (it stays warm the whole time the PC is on). The resistance is high, when NTCR1 is cold. You should wait 30 seconds after turning off a PC at the back, before flipping it back on at the back. So NTCR1 can be cold again, and limit the inrush. http://www.pavouk.org/hw/en_atxps.html We had a product at work, where NTCR1 started burning out on units in the field. I've not heard of that problem for a couple decades now, so I guess whoever designs supplies, figured that one out. That's one of the reasons I "harbor suspicions" about that way of doing it. Paul |
#9
|
|||
|
|||
Surprise!
In message , Paul
writes: [] So yes, a 1 second outage, the kind the power company likes to make when doing things in the switch yard, the PC can drop on one of those. You want a UPS to help you withstand 1 second outages. For tiny load glitches, the 200mS in that ATX capacitor can smooth things over. [] I find laptops/netbooks survive such things no problem (-:. Of course, using them on the mains all the time will/won't eventually knacker the battery, or at least severely reduce its capacity, depending who you ask and what sort of battery (and charging circuit) it is; however, for most people who use them on the mains all the time, the battery will retain enough capacity to certainly get over power glitches, and usually even to carry the PC to another room, for longer than the computer lasts. (I have _one_ elderly Toshiba [Compaq Armada 1750] where the battery has completely died, but that machine came [second-hand] with '98 installed, and I think actually predates that as it doesn't have Window keys. It is virtually unique among laptops that I've encountered in having the power supply built in, i. e. the mains lead [US: line cord] is just that, a lead; I'd far prefer that on any machine, but the craze for quoting the weight of laptops killed the idea. [Though with modern machines and power supplies they could re-address it.]) -- J. P. Gilliver. UMRA: 1960/1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf You can tell that you are getting old when typing VCR into a document [results that the spell checker] wants to change it to Vicar. - Brian Gaff in uk.tech.broadcast, 2016-3-6 |
#10
|
|||
|
|||
Surprise!
Paul wrote in news
KenK wrote: Paul wrote in news KenK wrote: I was innocently doing some Google searching with the Srware Iron browser and had just clicked on a Walmart site. While waiting for it to load I looked away to do something else and when I looked back the computer was booting! I vaguely recall a similar reboot many many years ago but not again for many years. Something to be concerned about or just a seldom- seen XP bug? If it matters using Kaspersky virus protection. XP Home with SP3 TIA Does the Kaspersky have a log ? Yes, but the three threat logs are all empty. So evidently the boot was not caused by a virus? I don't see any other applicable logs. Then you're off to Event Viewer for a look. Event Viewer? ******* Does the Walmart site allow third-party advertising ? Like, with a Flash exploit in it ? I don't know. Not familiar with Flask exploit. Paul -- I love a good meal! That's why I don't cook. |
#11
|
|||
|
|||
Surprise!
KenK wrote:
Event Viewer? Control Panel : Administrative Tools : Event Viewer When things break, there can be a log entry. If the error was an actual BSOD, and you have automatic restart enabled, then when you come back the machine might already have finished rebooting. On an actual BSOD, no item need be written into Event Viewer (because it crashed). It's on a more "controlled" error condition, that there is a possibility of a "hint" in the Event Viewer, somewhere. I don't know. Not familiar with Flash exploit. Flash is more than movies. It has a programming language. The programming language is probably why Flash keeps getting patches for security issues. Major websites, sometimes sell advertising space. It's a commodity kind of thing, kinda like selling cattle. It's sold in droves. Hardly anyone checks advertisements, to make sure there is nothing nasty in them. The "bad guys", they like to buy advertising on Yahoo, and gain access to the largest number of viewers possible. Then, they insert a Flash-based advertisement, that breaks out of any "box" put around the Flash plugin, and infects the machine. That's what the exploit is. It's an attack against the exposed surface of the Flash programming language. Just as it's recommended not to keep Java from Sun/Oracle installed on a computer (because it has a few holes in it), the same goes for Flash. This is why at least one browser, keeps track of the exploit-able state of Flash, and puts up a cross-hatched box instead of your movie, as a warning that Adobe has determined version X-1 is under attack, and you should update to version X to protect yourself. In your case, I expect it's the reverse situation. The latest version of PPAPI flash for Iron, has a bug in it. And it makes "ugly" movies on my machine. Parts of the video surface are messed up. It could be, that the latest Flash on your machine, is doing more than what happened on mine, and it's actually causing a crash. I can't remember the last time I had trouble like this with Flash. It's been good for a couple years at least. Flash is "a rock and a hard place". 1) Can't roll back to the last version, as it has a security hole. 2) No new version available... yet. 3) Can't remove it, because stupid sites still use it. My Canadian news web site still uses it. Paul |
#12
|
|||
|
|||
Surprise!
Paul wrote in news
KenK wrote: Event Viewer? Control Panel : Administrative Tools : Event Viewer When things break, there can be a log entry. I'm going to have to do some research to discover what the codes and names mean in these listings. However, I didn't see anything at the event time that was different from other times. Hebe to look more carefully. If the error was an actual BSOD, and you have automatic restart enabled, then when you come back the machine might already have finished rebooting. On an actual BSOD, no item need be written into Event Viewer (because it crashed). It's on a more "controlled" error condition, that there is a possibility of a "hint" in the Event Viewer, somewhere. I don't know. Not familiar with Flash exploit. Flash is more than movies. It has a programming language. The programming language is probably why Flash keeps getting patches for security issues. Major websites, sometimes sell advertising space. It's a commodity kind of thing, kinda like selling cattle. It's sold in droves. Hardly anyone checks advertisements, to make sure there is nothing nasty in them. The "bad guys", they like to buy advertising on Yahoo, and gain access to the largest number of viewers possible. Then, they insert a Flash-based advertisement, that breaks out of any "box" put around the Flash plugin, and infects the machine. That's what the exploit is. It's an attack against the exposed surface of the Flash programming language. Just as it's recommended not to keep Java from Sun/Oracle installed on a computer (because it has a few holes in it), the same goes for Flash. This is why at least one browser, keeps track of the exploit-able state of Flash, and puts up a cross-hatched box instead of your movie, as a warning that Adobe has determined version X-1 is under attack, and you should update to version X to protect yourself. In your case, I expect it's the reverse situation. The latest version of PPAPI flash for Iron, has a bug in it. And it makes "ugly" movies on my machine. Parts of the video surface are messed up. It could be, that the latest Flash on your machine, is doing more than what happened on mine, and it's actually causing a crash. I can't remember the last time I had trouble like this with Flash. It's been good for a couple years at least. Flash is "a rock and a hard place". 1) Can't roll back to the last version, as it has a security hole. 2) No new version available... yet. 3) Can't remove it, because stupid sites still use it. My Canadian news web site still uses it. Paul -- I love a good meal! That's why I don't cook. |
#13
|
|||
|
|||
Surprise!
VanguardLH wrote in :
KenK wrote: I was innocently doing some Google searching with the Srware Iron browser and had just clicked on a Walmart site. While waiting for it to load I looked away to do something else and when I looked back the computer was booting! I vaguely recall a similar reboot many many years ago but not again for many years. Something to be concerned about or just a seldom-seen XP bug? If it matters using Kaspersky virus protection. XP Home with SP3 Sometimes you may not notice power glitches (outages). They are short to you, maybe so short you won't see the lights flicker, especially if it is daytime and you have no lights turned on, but the outage duration is long enough that the capacitors in the power supply in the computer cannot cover the flicker in power. 300 ms is too long for a PSU but very short to you. Capacitors as well as everything else in a PSU will age and not perform as well over time. Also, brownouts can cause the PSU to shutdown and those can be momentary. Some UPSes will compensate for brownouts (for awhile but not indefinitely) by supplanting battery energy to what is obtained from the wall outlet. Some don't. Do you have a UPS to which your computer is attached? Yes. If so, either it is undersized or its battery is too old and needs to be replaced. If not, you're exposed to those power flickers. Most, maybe all, UPSes have software you install to control the power state of the computer and communicate to the UPS. The point of a UPS is to prevent slamming down the computer in a power outage but to shut it down gracefully. The UPS software will keep the computer up but only for so long depending on the remaining capacity and load on that is reported by the UPS. The software will perform a shutdown when there is not enough remaining run-time in the UPS. With an undersized UPS, its software client will perform an almost immediate shutdown of Windows during a power outage. It's always worked in the past but as you say, could be the batteries. However, the UPS warning lights have always signaled worn out batteries in the past. I've had it for a looong time. Some software for UPSes will mention when they are correcting for a power outage or brownout but not all do. If the remaining capacity for the UPS is too low, there won't be any uptime left on the UPS so the software has to perform an immediate graceful shutdown in an attempt to preserve data integrity. A right-sized UPS is for data protection. An oversized UPS is for convenience in keeping the gear up longer than needed for data protection; i.e., for you to keep using the computer. Alas, many users buy under-sized UPSes. Also, with age, the battery in the UPS will have less capacity, so one that started oversized may become undersized. I had a 2 kVA UPS that gave me 20 minutes of uptime after a power outage (would've been longer with a smaller load). After 5 years, that was down to around 2 minutes. I replaced the batteries (which constituted 60 pounds of the total 80 pounds of weight for the UPS) and got back up to the full uptime. While some UPSes have an alarm that sounds when the UPS is supplying power (whether for a full outage or to supplement during a brownout), not all do. Even for those with alarms, users may configure the UPS or the software to not sound off during a power event. The alarm on mine was very loud so I put some tape over the speaker to muffle it a bit. If you have a UPS, and if its software has logging and logging was enabled, check that log to take note of power events (outages, brownouts, etc). See if one of those power events coincides with when your computer did a shutdown. Either it didn't have software or I've not installed it. -- I love a good meal! That's why I don't cook. |
#14
|
|||
|
|||
Surprise!
KenK wrote:
If you have a UPS, and if its software has logging and logging was enabled, check that log to take note of power events (outages, brownouts, etc). See if one of those power events coincides with when your computer did a shutdown. Either it didn't have software or I've not installed it. Maybe you could check if the sofware is still around or maybe what they offer will still support your old UPS. The logging is helpful. It also is what provides a graceful shutdown of the computer instead of slamming it off when the UPS runs out of power during an outage; i.e., without the software, you've delayed when the computer gets slammed off. The software monitors the run-time reported by the UPS so it knows to perform a shutdown operation when there is, say, less than 2 to 5 minutes of run-time left (to provide sufficient time to do a graceful shutdown). Some UPSes are supported by Windows, like APC, but might not has all the same features (I've never setup a UPS using the Windows function). https://www.microsoft.com/resources/....mspx?mfr=true http://tenholder.net/tenWare2/tenHsP.../ApcUpsWin.pdf What old UPS brand and model do you have? Does it use USB or a serial port to connect to the PC? I've had some that used a serial port but you could not use a standard RS-232 cable. Instead the UPS maker switched some of the pins so you had to use their matching serial cable (one end was marked "UPS"). If you don't have any software installed for the UPS then the shutdown you experienced (where you saw a shutdown message and not just the computer slamming off) was not due to a power or UPS problem. |
#15
|
|||
|
|||
Surprise!
VanguardLH wrote in :
KenK wrote: If you have a UPS, and if its software has logging and logging was enabled, check that log to take note of power events (outages, brownouts, etc). See if one of those power events coincides with when your computer did a shutdown. Either it didn't have software or I've not installed it. Maybe you could check if the sofware is still around or maybe what they offer will still support your old UPS. The logging is helpful. It also is what provides a graceful shutdown of the computer instead of slamming it off when the UPS runs out of power during an outage; i.e., without the software, you've delayed when the computer gets slammed off. The software monitors the run-time reported by the UPS so it knows to perform a shutdown operation when there is, say, less than 2 to 5 minutes of run-time left (to provide sufficient time to do a graceful shutdown). Some UPSes are supported by Windows, like APC, but might not has all the same features (I've never setup a UPS using the Windows function). https://www.microsoft.com/resources/.../xp/all/proddo cs/en-us/pwrmn_ups_configure_ups.mspx?mfr=true http://tenholder.net/tenWare2/tenHsP.../ApcUpsWin.pdf What old UPS brand and model do you have? Emerson UPS 600 Does it use USB or a serial port to connect to the PC? No connection. Just the A/C power cord. I've had some that used a serial port but you could not use a standard RS-232 cable. Instead the UPS maker switched some of the pins so you had to use their matching serial cable (one end was marked "UPS"). If you don't have any software installed for the UPS then the shutdown you experienced (where you saw a shutdown message and not just the computer slamming off) was not due to a power or UPS problem. -- I love a good meal! That's why I don't cook. |
|
Thread Tools | |
Display Modes | |
|
|