« first day (1676 days earlier)      last day (2130 days later) » 

12:28 AM
[ SmokeDetector | MS ] Potentially bad keyword in answer: Can I get drunk by evaporating steam of alcohol? by edward murray on cooking.SE
naa- by WELZ
 
1:00 AM
Recovered from ssl.SSLError: [SSL: BAD_LENGTH] bad length (_ssl.c:2130)
[ SmokeDetector | MS ] Potentially bad keyword in answer: Why does Stanley Uris react so gruesomely to It returning? by toomanyparentheses on literature.SE (@Mithrandir)
 
1:18 AM
@tripleee yeah I don't think he replied any further after this twitter.com/quartatertots/status/996158657191591938
that was pretty much the peak anyways
Oh, I see Makyen already linked them all
 
@SmokeDetector ignore-
 
fp- by Makyen
[ SmokeDetector | MS ] Bad keyword in answer: BlackBerry - Play mp4 video from remote server by dave clark on stackoverflow.com
[ SmokeDetector | MS ] Blacklisted user: Botão direcionando para pagina visitada by Marcos Fernandes on pt.stackoverflow.com
[ SmokeDetector | MS ] Blacklisted user: Ocultar seta de navegação - Ionic by Marcos Fernandes on pt.stackoverflow.com
tpu- by Makyen
[ SmokeDetector | MS ] Blacklisted website in body: What is Bitcoin (virtual currency)? by PCBindex on bitcoin.SE
 
1:59 AM
sd k 3- 2k 2f 2k
 
2:17 AM
[ SmokeDetector | MS ] Potentially bad keyword in answer: Separate Java and HTML in jsp? by user3349977 on stackoverflow.com
 
@SmokeDetector 3 Answers promoting the same software in 13 minutes.
!!/watch thedroidhax\.com
 
@Makyen Added thedroidhax\.com to watchlist
[ SmokeDetector ] SmokeDetector started at rev cf02cee (SmokeDetector: Auto watch of thedroidhax\.com by Makyen --autopull) (running on teward/Solar Flare)
Restart: API quota is 19283.
 
@SmokeDetector f
 
2:56 AM
What are smokey queue sizes based on?
 
@thesecretmaster github.com/Charcoal-SE/SmokeDetector/blob/… is the logic for deciding when to fetch posts
 
coinflips
 
I'm not familiar with the code, but it looks like it's "when it has 2 or more items, except for special cases"
 
!!/apiquota
 
@thesecretmaster The current API quota remaining is 19111.
 
3:02 AM
Any reason that's 2 and not 1? I've pretty much never seen API quota below 12k.
 
it's 1 for low-traffic sites, since those are in special cases
 
default case 1 would be very bad I suspect
 
@thesecretmaster Looking at graphs, sometimes we get to ~8k
 
Ah, ok
Are there numbers on posts scanned per day anywhere?
 
3:08 AM
And that's just one instance, right?
 
Yep
But it's the most commonly-used instance
 
So possibly up to 150k
Well, there goes that idea
 
Those numbers probably include both answers and questions though -- we only make one request per question
So something like 30k-50k questions is probably a more realistic number
I could be wrong about that though
 
Still, > 20k
 
yeah, though many of them come from high-traffic sites, which are special-cased
 
3:13 AM
it's about 50k i think
 
3:29 AM
[ SmokeDetector | MS ] Potentially bad NS for domain in body: Front page duplicate images by crhatigan on stackoverflow.com
 
@quartata ummmm your tweets are psychedelic
 
fp- by tripleee
 
but VVVLQ, [tag:cv-pls]:ed to SOCVR
 
3:43 AM
What's the unit for scan rate?
Mposts/s? :)
 
Hertz
 
At 150kposts/day it would be about 1.7posts/s.
 
@SmokeDetector still up
 
[ SmokeDetector | MS ] Link at end of body: WPS Office not showing readonly notification on ubuntu? by Benin on askubuntu.com
fp- by tripleee
Restart: API quota is 18855.
 
@Makyen so it should be some x0.1 Hz ?
 
3:58 AM
I was just kidding, posts/s is the only metric which really makes sense
that comes out around 0.6Hz
 
0.6Hz = 0.6 p/s
 
1.7Hz would be 1.7posts/s. Hz is cycles per second. In this instance, posts would be considered cycles.
 
... sorry, need to get my morning coffee
 
That page shows a lot of numbers ranging from 10 to 20, so x0.1 Hz would be reasonable, right?
 
4:01 AM
I don't understand this question, though ... do you mean 10.1 or 20.1 (as in x is a digit) or some multiplier (as in x as a multiplication sign)?
 
According to this it's posts/s
 
@iBug No, 0.1Hz would be one every 10s, or 0.1post/s * (60s/1m) * (60m/1h) * (24h/1day) = 8,640 posts/day.
 
@tripleee oh, yes, sorry I forgot to say not to click onto my feed
I was on mobile so I couldn't get the direct image URL
 
made my morning (-:
 
OK, I meant *0.1Hz, which would be less ambiguous
 
4:07 AM
So it turns out the scan rate is not quite what we think it is (assuming y'all are thinking the same way I am)
 
[ SmokeDetector | MS ] Mostly dots in answer: "node run dev" can't work ,how to fix it? by francais on stackoverflow.com
 
the RSS hits in Charcoal Test look vaguely like something we could feed straight to Smoke Detector
 
fp- by tripleee
 
Scan rate is <number of posts scanned> divided by <time spent scanning>
Not <number of posts scanned> divided by <total time>
 
@tripleee those are surprisingly more comprehensible than the SE posts that advertise them
but not by much
@NobodyNada jeez, makes it surprising that it's so noisy then
 
4:09 AM
In other words, scan rate is an estimate of how many posts per second we're able to scan, not how many posts per second we are scanning
@quartata yeah, it doesn't even include API req time
 
@quartata yeah, I was thinking the same thing -- the least wacky spam posts here are probably more or less identical; perhaps we should look for duplication?
but anyway, having a bot subscribe to all of their RSS feeds and report any which don't get flagged would give us some sort of advance warning
though based on the ones I added yesterday, apparently not all of them actually end up in spam here
let's keep watching for a while still
 
@thesecretmaster There are multiple options for getting access to >20k SE API requests per day for SD. 3 off the top of my head: A) We could just ask SE to increase the quota in order to allow us to be more responsive (asking shouldn't hurt). B) We could switch instances when one instance runs out, as the quota is per IP. C) Delegate requests to currently active users, who then feed data to SD, but consuming the user's SE API quota.
NOTE: Each SE user (at 1 IP address) has up to 60k requests which can be used per day (50k reserved just for that SE user and 10k shared by all non-authenticated apps on that IP) (Yes, this means that authenticated bots using a bot account get a separate per user allocation, but user's shouldn't be using it that way). Thus, asking for, or just using multiple IPs to get, a larger total number of requests is reasonable, IMO.
 
@Makyen do we need more though?
 
We can reasonably consider that these SD accesses are just proxies for many people who could be implementing this functionality on their own, or each implementing a portion of it. Assuming that there are 20 consumers of SD reports which have available a total of 1.2M SE API requests/day. If you wanted to look at it as a trade, I'm sure we could find enough people who would be willing to give up 2k, 5k, 10k, or more of their 60k requests/day.
 
if we are in EC2 anyway scaling out to multiple instances with multiple IPs would not be very hard technically
but making the architecture depend on AWS is a limitation we might not want
 
4:15 AM
From a long term multiple site perspective, the queuing system should always exist in some capacity
 
@NobodyNada Well, we commonly see people mentioning that the spam or R/A post has been on the site for an extended period (at least 10's of minutes). I think it's reasonable to want to reduce that.
 
Would be useful for APIs that do quotas via x requests/minute
 
@Makyen That's a good point, although it might be feasible to prevent that by adding a cap to the max amount of time a post can stay in the queue (e.g. 10 mins)
I don't have stats in front of me though, just guessing
 
@quartata I agree that a queuing system should/needs to exist. However, it shouldn't be just fill to # posts, it should also timeout after X time if the queue for that site doesn't fill. That time could also be reasonably set based on what's expected from average traffic on that site/time/day of week.
 
Maybe we set a max amount of post-minutes
For example, with a threshold of 5 post-minutes, we fetch after 5 minutes if we have 1 post, 2.5 minutes for 2 posts, etc.
That specific case doesn't seem right as 5 minutes is short for one post and 1 minute is way too long for 5 posts; maybe something like posts-squared times minutes would be better
 
4:20 AM
@NobodyNada I'd suggest that we determine a maximum time and also look at what the average time is to fill a queue on various sites. We could, additionally, have it set so that it's below N standard deviations of the average (not dynamically calculated).
 
Restart: API quota is 18753.
 
!!/errorlogs 50
 
different sites might also have different read/write ratios ... slow sites with little traffic probably don't suffer a lot if a spam message is visible for 5 minutes because that translates to 0-1 visitors, whereas on some sites which have a lot of lurkers, 5 minutes might mean 0 new messages but many new visits to a post at the top of the start page
 
@tripleee right, so we'd want to adjust the thresholds per-site
 
@ArtOfCode flovis isn't catching exceptions in stage
 
4:23 AM
@NobodyNada Some formula like this would be reasonable, perhaps in combination with a normal queue limit, where we consider it full at N posts. Or in other words, I don't think we need to ditch what we've got, just add something that results in fewer long delays for posts.
 
sd poof
Why do all our sockets keep closing lmao
!!/location
 
@quartata teward/Solar Flare
Recovered from ssl.SSLError: [SSL: BAD_LENGTH] bad length (_ssl.c:2130)
 
@quartata Maybe Flovis network operations add to the noise though
 
Do we have the ability to get numbers for how long it takes the current queues to fill? average? daily? per hour?
 
We can get some numbers from the queue_timings script -- Smokey records the timings for every (?) post IIRC
Just ran it on RPi; here's the results:
 
4:26 AM
[ SmokeDetector | MS ] URL in title, bad NS for domain in body, bad NS for domain in title, bad keyword in body, blacklisted website in body, +5 more: reviverxtry.com/elemor-cream/ by robotalfart on english.SE
 
SITE,MIN,MAX,AVG,Q1,MEDIAN,Q3,STDDEV,COUNT,98P_MIN,98P_MAX
apple,0.001357,34.174321,0.4398282104072397,0.002112,0.00294,0.009357,2.2613915058436262,442,0.001357,4.962611222094492
computergraphics,0.001787,3212920.86682,807172.395435375,0.007173,10193.409104,3212916.399257,1388975.9514405213,8,0.001787,3212920.86682
arduino,0.001035,118371.256783,749.7732819810122,0.001888,0.003274,0.01285,9387.216726405355,158,0.001035,19524.206734791722
startups,0.003283,0.013434,0.006717,0.003283,0.003434,0.013434,0.0047500362805631986,3,0.003283,0.013434
 
tpu- by NobodyNada
 
IIRC those are stats for how long posts tend to sit in the queue
We could probably get better numbers from Solar Flare, but I don't have access to those numbers since I'm not Thomas
 
WIBNI these were reported somewhere central
 
@tripleee WIBNI?
 
@tripleee thanks -- I tried a three-finger tap (which searches dictionary and Wikipedia on Mac), but was too lazy to actually Google it, sorry about that
 
sure (-:
 
I'm probably too tired to be chatting right now :p
as evidenced by the fact that I keep spitballing facts in the hope that someone else finds them useful
 
looking some more at the RSS hits, Zdorov Propolis (which I watched yesterday) was RSSed on March 20 and appeared in spam on March 20th-21st but there were actually some hits before then, it's actually blacklistable now but it's probably past its useful life now anyway ... guessing it will have appeared in earlier RSS posts as well though
(the RSS feeds show some old posts when you initially set them up, so that's why it came up yesterday)
I don't think Halflife wants to be an RSS reader, this should arguably run in a separate bot, though I guess it might make sense to make Halflife more of a plug-in host and allow things like this to be added there without significant code changes if ideas like this keep coming up
 
[ SmokeDetector | MS ] Potentially bad NS for domain in answer: How to change "DWG Association" setting on AutoCAD? by Tomas on superuser.com
fp- by tripleee
 
4:36 AM
hmmm, first FP for the Siteground watch I added yesterday?
 
might be NAA rather than FP
yeah, link-only
 
naa by NobodyNada
naa- by Makyen
 
agreed
still iffy NS does not automatically mean spam site, just cheap site
but let's keep it around for a little while longer still to see whether it can possibly contribute some value
 
[ SmokeDetector | MS ] Bad NS for domain in body, bad keyword in body, blacklisted website in body, pattern-matching website in body: Tips For Losing Weight by Toneeloy on puzzling.SE (@Mithrandir)
tpu- by NobodyNada
 
good old Lloyd+Reza continues to work fabulously ^
 
4:43 AM
fp- by NobodyNada
fp by Nisse Engström on Ocultar seta de navegação - Ionic [MS]
fp by Nisse Engström
 
@SmokeDetector I do wish we could tag questions as "vandalism/VLQ" when they are VLQ
(that's the label in FIRE, I guess it's different in Metasmoke)
 
5:03 AM
[ SmokeDetector | MS ] URL in title, bad NS for domain in body, bad NS for domain in title, bad keyword in body, bad keyword in title, +4 more: www.usahealthmarket.info/spectrum-cbd-hemp-oil/ by user122783 on graphicdesign.SE
tpu- by NobodyNada
 
@NobodyNada What are the units? Given that the min and max can be 10 orders of magnitude apart, I'm not sure what I should use. I'm assuming that these are in seconds. However, using seconds results in the maximum average delay (per site) being 4.6 days for sites with > 100 posts, and 2.9 days for sites with > 1000 posts.
 
@Makyen I have no idea, but I'd guess seconds. The skips are probably because of my Pi being in standby mode most of the time
IIRC the queue is saved across reboots, so it saves a post ID, goes into standby, and scans the posts again when it wakes up potentially months later
That's why Solar Flare will give you better numbers
 
OK. I'm reasonably comfortable with considering the data to be flawed, at least for sites with < ~2k posts. Unfortunately, an SD instance collecting this data across periods of time when it's in standby, or reboots that last extended periods of time, makes the data less useful. While we do want to know about the effect of those overall, it really only makes sense to include them in the data when the instance is the one which is intended to be actively processing posts throughout the downtime.
 
5:21 AM
if you have Python, you can run your own instance for a while, there's an option to not have it connect to any chatrooms and it won't even try to talk to Metasmoke if you don't supply an MS key
if I get the time to proceed with the phone numbers refactoring I'll probably kick up one in Charcoal Test anyway
but that will be next week earliest (and if not next week then probably not before August)
 
[ SmokeDetector | MS ] URL in title, bad NS for domain in body, bad NS for domain in title, bad keyword in body, bad keyword in title, +4 more: gomusclebuilding.com/vital-x9/ by InezSmith on puzzling.SE (@Mithrandir)
tpu- by Makyen
 
A J
@angussidney you're here?
 
@AJ I'm on mobile at the moment
Regarding your issue: does it only happen while Pycharm is open?
 
A J
@angussidney When I open PyCharm and it is loading projects. I also have problem accessing data residing on the disk PyCharm is installed on.
 
That's...weird
 
5:28 AM
[ SmokeDetector | MS ] URL in title, bad keyword in body, bad keyword in title, blacklisted website in body, blacklisted website in title, +3 more: advancemenpower.com/climadex/ by ffertsssh on astronomy.SE
 
Did you install directly or via the toolbox app?
 
tpu- by Makyen
 
A J
toolbox app? I downloaded the software from their official site and then installed it.
 
@AJ do you have any problems at all while not using PyCharm?
 
A J
@NobodyNada Yes. This one time when I was using Python CLI.
 
5:32 AM
Try checking whether there are any problems with your hard drive: howtogeek.com/134735/how-to-see-if-your-hard-drive-is-dying
 
@AJ interesting. That's still Python-related, but the weird thing is that it's somehow able to impact other applications and causes blue-screens
 
It's possible that there is something wrong with your hard drive, the Pycharm happened to be the data which was written over it.
Causing some sort of issue
 
A J
@NobodyNada I'll try to take a picture of the crash message next time if it happens.
 
@angussidney I'd suspect something along those lines
 
A J
@angussidney Oh.
Thanks for the link. I will check it once I get home.
A query.
 
5:35 AM
Either Python has an insane privilege level, and is malfunctioning...or Windows is malfunctioning in a way that is somehow triggered by Python
 
I wouldn't be surprised at all if the hard drive has gone dodgy after 7 years
 
I think some sort of hard disk problem seems like the most likely explanation
 
@tripleee Don't rush to do it on my account. It'd be interesting to know, but isn't critical.
However, I do feel it would be good to set some upper limit on the delay in the queue, just so it's bounded. Having the data would just let us get a feel for what the impact on quota might be, but there's nothing that wrong with just proceeding with setting a max queue delay at some larger number, say 10 minutes, and seeing how it impacts remaining quota. Worst case is that we run out of quota on an instance and roll the change back.
 
A J
Suppose there is something wrong with the hard drive, I have to replace the hard drive. Right?
 
@AJ Most likely
I'd recommend putting an SSD in, you'll get a huge performance & reliability boost
 
5:37 AM
^^^^^^^
100%
 
A J
@NobodyNada Then I have to reinstall the OS?
 
I will never buy a computer without an SSD ever again
2
 
@AJ You could just clone your old drive, though if the data is corrupt then problems could carry over
 
A J
@angussidney Isn't SSD more expensive than HDD?
 
@angussidney Same, although mine didn't have a huge impact, it was noticeable enough for me to make a difference.
 
A J
5:39 AM
@NobodyNada Actually the data is not corrupt. Not yet.
 
@AJ Yes
 
@angussidney A friend has a 10-year-old MacBook that is usable as a brand new computer, since he replaced the HDD with an SSD
 
tpu- by tripleee
 
@AJ that's...hard to prove
 
5:39 AM
@AJ it depends, is the problem drive the boot drive or a secondary drive?
 
!!/watch mg\W?infomatics
 
@Glorfindel Added mg\W?infomatics to watchlist
 
A J
@angussidney I really don't know. I have Windows installed in C drive and there are two more drives.
 
@AJ yes, that's the only downside. That's why in my PC I have 250GB ssd for booting, and a cheap 1TB hard drive for mass storage
 
CI on f89e8e2 succeeded. Message contains 'autopull', pulling...
 
5:41 AM
@AJ which drive did you install Pycharm onto? Also which drive did you say was having issues with accessing files?
 
Restart: API quota is 18325.
 
Data corruption isn't necessarily going to be easy to spot. If it's really bad, it'll be obvious, but if not the symptoms could be very obscure or minor
 
A J
@angussidney The other drive. And this is the drive where I have problem accessing files.
 
@AJ OK. Then that means that you might be able to get away without having to reinstall windows
 
@AJ this == the other drive?
 
A J
5:42 AM
@NobodyNada I though have an external HDD where I keep backup.
@NobodyNada yes
 
if I were you, I'd try installing PyCharm on another drive and see what happens
 
maybe call them C: and D: for easier reference even if you call them something else locally
 
[ SmokeDetector | MS ] Bad NS for domain in body, bad NS for domain in title, bad keyword in body, bad keyword in title, bad pattern in URL body, +5 more: Konect Nutra Keto www.muscle4supplement.com/konect-nutra-keto-reviews/ by Matthenfr on askubuntu.com
 
if windows is on C: and the problematic drive is D: then replacing D: will not require a reinstall of the OS
 
tpu- by tripleee
 
5:45 AM
(... though for other reasons replacing Windows with another OS is always a thing I like to recommend)
4
 
A J
I have three partitions. C, D and U (I chose this letter because I had installed Ubuntu on it long back ago). Now, I have Windows on C and PyCharm on U.
 
@tripleee yeah, I got my sister to switch to Ubuntu
Way more performant than Windows, though sometimes requires a bit more troubleshooting
 
A J
@NobodyNada Windows is still first choice for personal computers in India. People here find Windows easier to use than Ubuntu.
 
that's weird, here too but it's not really
 
@AJ yeah, it's like that in the rest of the world as far as I know
 
5:48 AM
the one thing Windows has which is hard to top is popular commercial software and perhaps especially games
 
I think it's probably more of a familiarity thing
 
[ SmokeDetector | MS ] Potentially bad keyword in body: Can't Access Local Static Resource Flask Application Deployed on Apache2 by webguybit on stackoverflow.com
[ SmokeDetector | MS ] Bad keyword in body, bad keyword in title, blacklisted website in body, body starts with title and ends in URL, link at end of body: Total Tone Consuming wholesome is even extra by kigeli on askubuntu.com
 
@SmokeDetector SOCVR
 
A J
@NobodyNada Yeah. When you first learn how to use a computer, they will give you a tutorial on Windows.
 
5:51 AM
[ SmokeDetector | MS ] Blacklisted website in body: Where can i get Brigade Woods Brochure? by Dream Earth on stackoverflow.com
tpu- by Makyen
 
!!/watch setup-office\.com
 
@tripleee yeah. I dual boot Windows + Linux on both my laptop and desktop. On the laptop I prefer Linux, since I mainly use it for development and elementaryOS runs faster than windows, but on my PC I stick with Windows as my primary OS due to the game compatibility
 
@NobodyNada Added setup-office\.com to watchlist
 
!!/watch brigadewoods\.net\.in
 
@SmokeDetector \o/ Misleading link did it's job :-)
 
5:52 AM
@NobodyNada Added brigadewoods\.net\.in to watchlist
 
!!/watch brigade\Wwoods
 
@NobodyNada Added brigade\Wwoods to watchlist
[ SmokeDetector | MS ] Link at end of answer: How to deobfuscate a HTML document by Dipesh Patil on stackoverflow.com
Restart: API quota is 18244.
[ SmokeDetector | MS ] Bad keyword in body, bad keyword in title, blacklisted website in body, blacklisted website in title, link at end of body, +2 more: Official Site:-supplement4fitness.com/rapid-tone-diet-ca/ by jozyhoy on askubuntu.com
tpu- by Makyen
 
A J
I heard that it is not easy to recover data from SSD.
 
@AJ if it's damaged, you mean?
1) you're far less likely to have an SSD fail than an HDD, and 2) you should be making backups anyway ;)
 
A J
@NobodyNada lol ok
Should I run chkdsk to check if there are any problems and recover from it?
 
6:01 AM
Recovered from websocket._exceptions.WebSocketConnectionClosedException: socket is already closed.
[ SmokeDetector | MS ] Bad keyword in title: Which is the best SEO Company in delhi? by chamksam on askubuntu.com
 
@SmokeDetector Probably a spam seed
 
[ SmokeDetector | MS ] Link at end of body, pattern-matching website in body, repeated URL at end of long post: Nerve Renew all natural pain relief by maryramos on meta.SE
tpu- by PeterJ
tpu- by PeterJ
 
A J
> Technical support at Norton antivirus
 
@AJ where?
 
A J
chamksam, United States Virgin Islands
1
 
6:08 AM
Very suspicious
 
[ SmokeDetector | MS ] URL in title, bad NS for domain in body, bad NS for domain in title, bad keyword in body, bad keyword in title, +6 more: www.nutritionfit.org/total-tone-diet/ by Pande1949 on drupal.SE
 
A J
Of course.
 
tpu- by Makyen
 
A J
@SmokeDetector k
So, should I run chkdsk command to check for any disk errors?
 
@AJ As someone who's recovered data from failing HDDs on multiple occasions, I'd far rather have my data on an SSD than an HDD. But, as @NobodyNada's mentioned, you should be making backups of data you don't want to loose. Making backups on a regular basis, automatically, is something that's quite easy at this point.
 
6:10 AM
[ SmokeDetector | MS ] Bad keyword in answer, blacklisted website in answer: Which is a safe way to record Firefox passwords in Gnome Seahorse Keyring? by huahua on security.SE
tpu- by Makyen
 
A J
sd k
 
@AJ It shouldn't hurt, but it will only alert you to inconsistencies in the filesystem structure, not hardware problems with the drive.
 
A J
@Makyen Oh.
 
Hardware problems can cause filesystem corruption, but that's only a symptom, not the problem.
 
A J
@NobodyNada so, check for defragmentation?
 
6:12 AM
@AJ no, that won't help with anything
Defragmentation searches for files that are scattered across the drive and consolidates them
A fragmented drive is slower, but shouldn't have any errors
 
@AJ It depends on what you think the problem is and the severity of the problem.
 
A J
Ah. Okay.
 
@AJ No, that can only hurt.
 
@SmokeDetector title is sooo spammy, I'm going with spam even if there is no overt promotion
 
A J
6:14 AM
Hmm. Lemme find the problems using the link Angus gave me earlier first, then I will let you guys know if there is certain issue.
 
@AJ Sounds like a good plan
 
A J
Thanks.
 
Restart: API quota is 18093.
 
A J
I invested around 7.5k for battery and screen change in this laptop till now. And funny I purchased it for 33k.
 
anyway, time for bed over here. See you tomorrow
 
A J
6:18 AM
@NobodyNada g'night.
 
night o/
 
[ SmokeDetector | MS ] Bad NS for domain in body, bad keyword in title, blacklisted website in body, link at end of body, potentially bad keyword in body, +1 more: Testo E-Force TESTOSTERONE BOOSTER SUPPLEMENT by cliffewski5 on drupal.SE
 
@AJ Basically, you have two major classes of issues: A) Logical problems. These are OS/software based, where the logical information on the drive has become corrupted. Windows can cause these types of problems. If exclusively this class of problem, then the underlying hard disk hardware is fine. B) Hardware problem. The hard disk is failing and will require replacement. This can also be a cause of (A).
 
tpu- by Makyen
 
A J
Can anyone walk me through the term "6.0gb/s" for HDDs?
@Makyen Yeah, I am not sure at the moment. I will keep my phone nearby next time to take the picture of crash.
 
6:22 AM
transfer speed, doesn't sounds like a realistic speed though
 
A J
The thing is whenever it crashes, it automatically reboots.
 
if you can find a S.M.A.R.T. utility some drives have logic to report hardware problems through that
 
What to do really depends on if the data on the drive is important to you. If it is, the the first step is always to make a complete byte-for-byte backup of the drive and then perform any recovery operations on the backup (and if possible, I usually make a backup of the backup (not a second copy of the original, but a copy of the backup)).
 
yes, get a disk image now before you poke anything
 
[ SmokeDetector | MS ] URL-only title, bad NS for domain in body, bad NS for domain in title, bad keyword in body, bad keyword in title, +5 more: www.nutritionfit.org/total-tone-diet/ by Untioutilly on askubuntu.com
tpu- by tripleee
 
6:25 AM
!!/watch biopharmchem\.com
 
@tripleee That pattern looks like it's already caught by Blacklisted website in answer and Blacklisted website in body; append -force if you really want to do that.
 
I'll leave it to someone with more sleep than me to demonstrate how to fire the dd cannon
 
A J
@Makyen I don't have that much data to back up. mostly is my code and other personal stuff.
 
@SmokeDetector hmmm, only 1/1 and I don't see an explicit pattern for this
@AJ byte-for-byte includes the empty sectors
 
[ SmokeDetector | MS ] Username similar to website in answer: What's the relationship between Dalvik and Zygote process? by Languoguang on stackoverflow.com
 
A J
6:27 AM
@tripleee How do I backup empty sectors?
 
fp- by tripleee
 
@AJ the sectors are just sectors, dd reads them one by one, it doesn't know whether they are used for anything or not
 
@AJ My personal preference for making such byte-for-byte backup of a drive that is not known-good is ddrescue, a free Linux/Unix (also works under Cygwin on Windows) based program.
 
so typically they are filled with zeros or a filler pattern, but every sector on a formatted disk contains something
 
6:30 AM
oh wait, stupid, I had a typo in my grep, I blacklisted it myself in February
 
JAD
heh, I first read that as "I blacklisted myself in February"
 
@AJ this is in rupees?
 
A J
@quartata Yes.
 
I was wondering what kind of import duties you had if it was 7K USD
 
it's 5/5 now, they did 4/4 with www.biopharmchem.com over the course of 8 months last year starting in January
 
A J
6:33 AM
@quartata Yeah, I'd buy 2-3 Alienware with that money instead.
 
I found a nice SSD vs HDD comparison chart on Computer World but the site is so horribly advertising-infested that I don't want to link to it
 
!!/unwatch stellarinfo\.com
0/6
 
@Mithrandir Removed stellarinfo\.com from watchlist
 
... 'the research showed that a large proportion (56%) of the failed drives failed without recording any count in the "four strong S.M.A.R.T. warnings" identified as scan errors, reallocation count, offline reallocation and probational count. Further, 36% of failed drives did so without recording any S.M.A.R.T. error at all, except the temperature, meaning that S.M.A.R.T. data alone was of limited usefulness in anticipating failures.'
 
6:40 AM
Restart: API quota is 17900.
 
@SmokeDetector ignore
 
@Mithrandir Post ignored; alerts about it will no longer be posted.
tp by Mithrandir on Hardware-Based Password Manager [MS]
fp by Mithrandir
[ SmokeDetector | MS ] Username similar to website in answer: How can I generate apk file for create-react-native-app (using EXPO.IO component) by Johncy on stackoverflow.com
fp- by DavidPostill
[ SmokeDetector | MS ] Potentially bad keyword in answer: How to import wordpress posts with images and featured images? by amarendra sharma on stackoverflow.com
tpu- by PeterJ
tpu- by Makyen
[ SmokeDetector | MS ] Blacklisted website in body: Who promote your gigs on Fiverr? by Amir Shahzad on stackoverflow.com
tpu- by Makyen
 
6:56 AM
!!/watch seair\.co\.in
 
@tripleee That pattern looks like it's already caught by Potentially bad keyword in body and Potentially bad keyword in answer; append -force if you really want to do that.
 
!!/watch amirtopseller
 
@tripleee Added amirtopseller to watchlist
[ SmokeDetector | MS ] Bad keyword in body, bad keyword in title, bad keyword in username: What is keto slim? by keto slim on askubuntu.com
[ SmokeDetector ] SmokeDetector started at rev 54dfda4 (SmokeDetector: Auto watch of amirtopseller by tripleee --autopull) (running on teward/Solar Flare)
Restart: API quota is 17766.
 
@SmokeDetector k
 

« first day (1676 days earlier)      last day (2130 days later) »