Inconsistent download of ATIH2014_6614_en-US. Is it just me?
Folks:
This is very strange. I am downloading the ATIH2014_6614_en-US.exe file from "https://www.acronis.com/my/products/" and Internet Explorer 11 reports that the signature is corrupt or invalid (see attached JPG). Sure enough, if I check the MD5 hash (using HashTab), the hash does not correspond to the B4B10... posted on the product download page. So I decide to try again. I rename the file I just downloaded and restart the download freshly. I get the same IE warning again and the newly downloaded file reports a MD5 hash that again is different from the posted one. What is even more strange is that the two files I just downloaded differ from one another in the MD5 hash. Of course, the files have exactly the same size (number of bytes). I tried the same experiment from Firefox. While Firefox has no idea about corrupted signatures and shows no warning that something may be wrong, when I check the MD5 hash of the ATIH2014_6614 downloaded with Firefox, it returns yet further different numbers from all of the files downloaded so far!
I ran one more experiment: I went to the "Download older build page" where I can also access the 6614 build, and tried to download from there. Again, IE warns about the corrupted signature, and the hash of the downloaded file is different from all other ATIH2014 downloads and different from Acronis' page!
I cannot understand what is happening. I have a great Internet connection in my home and I never experienced problems downloading anything.
@Acronis - is there an FTP site I could try my download from, as a test?
@Forum Folks - could anybody try to download the 6614 build again and report on the MD5 success?
Many thanks for any help. I do not dare to attempt installation unless I get the correct MD5 hash on my download. Thanks to Acronis for posting it. It's the professional thing to do.
Les, thanks for trying this out. I have no idea what could possibly be the issue. Maybe my Internet connection is noisy although I'm sure I would have seen many more effects of that (delays, garbled VoIP, etc).
To answer your question, no, I just have the default Windows 7 firewall enabled. I have Norton AV 20.4.0.40. Hmm... Maybe Norton AV does some file imprinting during its AV check? I will shut it down and try again and then I will report here.
Thanks again to trying the download. At least this way I know it's a problem at my end.
Let us know how it comes out. It's always scary to disable your antivirus, but if there's a chance it is mangling a download, sometimes you have to do it, especially if you're careful and have a checksum you can verify the download with. Just be sure to turn it back on. :-O
Les, it does make sense that it could be the AV. I did my best to shut it down but, the annoying thing is, there is no such thing as an OFF button for Norton. They have 10 little "disable" sliders which I turned off. I then went into my Task manager and attempted to kill all programs and services associated with NAV. I was only partially successful there. There was one last service that could not be shut down. After I did all this I tried again and I got the same bad result - a completely new MD5! It could still be NAV but I don't know. I rebooted my system so I can get all the services running correctly again.
I am now suspicious of my internet connection. I am doing an extended ping to google and to my provider, Time Warner, and in both cases, every 30-60 seconds or so, I get a lost packet. I may just have an intermittent connection. I am placing a service call with those guys, just in case. I'll get to the bottom of this!
Meanwhile, I called my brother in Oregon and asked him to download it for me and check the MD5. No surprise there - he's is perfect. I then sFTPed that file from his machine to mine and I got it with perfect MD5. If I'm not mistaken, sFTP has some checksum error detection and retransmission protocol and that may have recovered the occasional missing bit. Bottom line, I have the file, even if through a very tortuous way.
Good to hear that you were able to get it one way or another. I'm surprised surprised there isn't some sort of other level error correction going on that prevents this sort of thing and allows for re-transmission using http (i.e. TCP is supposed to re-transmit bad packets, isn't it?). Reminds me of the days of ZMODEM.