March 2017 - Roundtable

Ryzen is here
by Trevor Cordes 23 Mar '17

23 Mar '17

My suppliers just informed me the new AMD Ryzen stuff is in stock. I'm doing quick glances at the choices now. First impressions: 1. Ryzen 7-series CPUs aren't cheap! Prices I am seeing will be just a bit below $500 to $700, though these may not represent all the models. That means cost savings over Xeon might not be substantial (or existent) at the low/mid-end. 2. ECC/board weirdness: I'm already confused, the high-end Asus X370 board doesn't support ECC but the B350 at half the price does? That doesn't instill confidence. I'm doing more research and when I get some final comparisons down I'll post them here. Sadly, it looks like any way I do an upgrade it won't be cheap. Anyone else doing the same research (esp with regards to ECC) post your findings here!

1 0

Winkler Linux User Group
by John Funk 20 Mar '17

20 Mar '17

Winkler Linux User Group Tomorrow Tuesday 2017-03-21 we will be meeting at 7:15 at the Winkler branch of South Central Regional Library. Please let anyone you know who might interested.The topic for the next meeting will be discussing security concerns and permissions with file servers. This will not be a heavy discussion.

1 0

GnuCash crash on secret payload in XML file
by Hartmut W Sager 14 Mar '17

14 Mar '17

OK, it's late at night, and I did something that Trevor likes doing. :) I went looking for trouble. But hey, I had a good reason, and had been planning this experiment for weeks already. My current accounting is still my own FoxPro concoction (started in 1992 and updated a few hundred times), and for the last few years I knew I would have to move it. Besides export/import issues, I have several extra fields of information that I need to preserve (even if not accessible outside of raw XML editing), so I edited a good GnuCash native XML file by adding a secret payload in one existing transaction with <mynewtag>mypayload</mynewtag>, at the same level as the other fields of that transaction. I was hoping it would be ignored on load, and restored on save. The result: GnuCrash crashed totally while opening the edited XML file! Yeah, just crashed with a crash message and self-quit after that. Is GnuCash that unstable??? BTW, a simple raw XML edit of an existing text field of a transaction works fine, and is then shown in GnuCash when the file is opened. Hartmut W Sager - Tel +1-204-339-8331

4 11

Suggestion For Improvements To Cryptographic Hashing
by Ihor Jakowec 14 Mar '17

14 Mar '17

I hope this is readable, if not please notify me. This article is also readable from google+ Keywords: Ihor Jakowec DEJA VU HASH Revised Or, copy and paste, the below link... https://drive.google.com/open?id=0BxLingfkcfkbQ3JwYUlZVVBGZGc DEJA VU HASH Abstract Described here is a self modifying cryptographic hash, that changes with every login, and/or, after an arbitrary time interval. In addition to a password, the hash uses login statistics. A hacker would would have less time to crack a password, because the hash value could keep changing. Also, hacker would need login statistics.  TIME SENSITIVE CRYPTOGRAPHIC HASH This method is used for remote logins from computers with an operating system containing log files that hold login data and network traffic data. This method is NOT for use by “dumb” terminals, or on guest computers where data is erased upon logout. The input to the hash consists of some or all of the following: 1.) Password. 2.) Time delay, established by use either/or: - the mode and variance of several ping attempts - a truncated value common to most ping trials - or the intersection algorithm. [1] 3.) Login time. 4.) Previous or, current login duration. 5.) CPU serial number. 6.) An array that is a histogram of values of the number of packets sent and received (the interval size is arbitrary) [1] _________________ Wikipedia Contributors “Intersection Algorithm” “Wikipedia, The Free Encyclopedia” Wikipedia, The Free Encyclopedia 5 May 2014. Web. 5 May 2014. NOTE: You could use NTP (Network Time Protocol), with peer-to-peer connectivity for both the login client and the server; to increase the resolution of the above items: 2, 3, and 4. (Referring to item 4.) If the current login is used, the hash is performed at the end of the login. A different hash value can be kept for each login from a different CPU. (This could be a weakness if many different CPU’s have been used, and infrequently at that.) Therefore, using item 5. should only be considered as an option, for those who want to restrict logins to a select few computers. How these values are arranged as input to the hash is arbitrary. You could use sequential concatenation. Or, values can be hashed separately, then xored to the final hash value. Input values can also be bit shuffled. Modification: Item 6 could consist of a single array that is a histogram of differences: (number of data packets sent - number of data packets received) The arbitrary time interval is: login duration / inter login interval Both the server and the client would have records of items 1 to 6 on their respective file systems. However, for concealment, the way they are stored, and used as input to the hash can vary. Since, this method results in a different hash for every login. A hacker’s cracking time would be limited to the time interval between successive logins PSEUDO LOGIN The concept here is to use pseudo logins. This is not a full login. Only the login state tables are synchronized and the password hash value is re-hashed. The login interval can be based on information known to both the client and the server. The average of the last five logins could be used, where a fraction, or multiple, of this average is used as the pseudo login interval. Moreover, the pseudo login interval can be vary. This interval can be made slightly longer or shorter. This variation can be based on the histogram of differences previously mentioned. The histogram can be partitioned into percentiles, or an arbitrary n-tile. Lower than average values can be used to shrink the pseudo login interval, while higher than average values can be used to lengthen the interval. Optimally, the average of these intervals should be slightly shorter than the amount of time needed by the average hacker to crack a hash value. Or, it could be left up to the system adminstrator to set the length of the pseudo login interval. This would depend on the need for security in juxtaposition to system load demands. SALTING THE SALT As an added security feature, each pseudo login hash can be salted with a random number. The same value that is used by the server and the remote login. The random number generator and the seed value used, would be common to the server and the login client. The type of random number generator used would be quasi-propriatary to the system being used. That is, each random value could be cryptographically hashed with a common seed value. The resulting hash value could be used with the hash value of the previous login or previous pseudo login hash value. ================================================== DISCLAIMER Any words not defined, in this collection of documents; use the definition given to them, by the O.E.D. and/or American Webster's dictionary. It is not my intention to use double entendre: either from English or in any other language, or from pop subculture (past or present). Any words redefined by me, or phrases defined by me, are simply technical in nature. They are not intended to refer to any: group, institution, organization, person or persons. — No alternate meaning to any word or phrase is implied or intended. — My intent is NOT to: slight, insult, affront, or offend. Ihor Jakowec Tuesday 14 December 2017 ===================================================

1 0

Re: [RndTbl] GnuCash crash on secret payload in XML file
by Dan Martin 13 Mar '17

13 Mar '17

I have used GnuCash for many years, building up huge XML files, and never had an issue until I tried some auto-entry with a program. If the file is not EXACTLY what it expects, it will fail. No problem getting data out, though it can be processor intensive to rebuild the trees. But I have not had the patience to dig into source code to be able to write to it. It would be nice to have a high level API to allow auto entry, modify reports, etc. -Dan On Mon, Mar 13, 2017 at 3:04 AM, Hartmut W Sager <hwsager(a)marityme.net> wrote: > OK, it's late at night, and I did something that Trevor likes doing. :) > I went looking for trouble. But hey, I had a good reason, and had been > planning this experiment for weeks already. > > My current accounting is still my own FoxPro concoction (started in 1992 > and updated a few hundred times), and for the last few years I knew I would > have to move it. Besides export/import issues, I have several extra fields > of information that I need to preserve (even if not accessible outside of > raw XML editing), so I edited a good GnuCash native XML file by adding a > secret payload in one existing transaction with > <mynewtag>mypayload</mynewtag>, at the same level as the other fields of > that transaction. I was hoping it would be ignored on load, and restored > on save. > > The result: GnuCrash crashed totally while opening the edited XML file! > Yeah, just crashed with a crash message and self-quit after that. Is > GnuCash that unstable??? > > BTW, a simple raw XML edit of an existing text field of a transaction > works fine, and is then shown in GnuCash when the file is opened. > > Hartmut W Sager - Tel +1-204-339-8331 <(204)%20339-8331> > > > _______________________________________________ > Roundtable mailing list > Roundtable(a)muug.ca > https://muug.ca/mailman/listinfo/roundtable > >

1 0

Les.net MUUG member discount slow in coming
by Hartmut W Sager 13 Mar '17

13 Mar '17

Sorry Les, no offense intended, but 4.5 months since the first promise is a bit long, and I did ask your staff about this a while back. The facts: 1. In the Nov 2016 MUUG Newsletter, Les.net offered a 10% discount to MUUG members. 2. My inquiry about this at Les.net a few weeks later led to a pleasant conversation with Sonja, who assured me that it would be implemented "pretty soon", and that discounting would definitely be back-dated to 01 Nov 2016. So, Les, will you kindly carry through with this? It was at your own discretion to offer this in the first place, without any pressure from anywhere (I hope) - I am not your typical Winnipeger who is routinely looking for a discount everywhere. I hasten to add, I too appreciate what you do for MUUG. Hartmut W Sager - Tel +1-204-339-8331

1 1

Looking for Proxmox support
by Gerald Brandt 10 Mar '17

10 Mar '17

Hi, I used to do consulting (about a year ago). When I stopped, I told my clients, and all was well. Well, I have an old client that called me up. He had serious issues and didn't know why. I solved the issues for him, but we are now looking for someone to move forward to continue support. The client is relatively low maintenance, but when he needs help, he needs it immediately. The configuration is 4 Proxmox 4.4 servers and two Synologys in HA configuration serving vm images under NFSv4. Is there anyone in Winnipeg that would be interested in supporting infrastructure like this? Gerald

1 0

MUUG Meeting, Mar 14, 7:30pm -- Pi-hole and Shred
by Gilbert E. Detillieux 07 Mar '17

07 Mar '17

*** Please note the updated meeting location below *** The Manitoba UNIX User Group (MUUG) will be holding its next monthly meeting on Tuesday, March 14 (Pi Day). The meeting topic for this month is as follows: Daemon-Dash: Pi-hole Pi-hole turns your Raspberri Pi (or other supported devices) into a network-wide ad blocker, using DNS redirection. It promises to block over 100,000 ad-serving domains, improve overall network performance and reduce cellular data usage, on any device. Gilbert Detillieux will demonstrate the installation process and typical use of Pi-hole on a Raspberry Pi (after all, it is Pi Day!) and describe some pros and cons of the installation method used, as well as some pros and cons of using an ad blocker on your network. RTFM: shred This month's RTFM will be presented by Wyatt Zacharias, who will talk about the shred(1) command, which allows you to overwrite a file to hide its contents, and optionally delete it. The group usually holds its meetings at 7:30pm on the second Tuesday of every month from September to June. (There are no meetings in July and August.) Meetings are open to the general public; you don't have to be a MUUG member to attend. ****************************************************************** Please note our *NEW* meeting location for this month: *Room 1L11* Lockhart Hall, University of Winnipeg, entrance on Ellice Ave. between Spence St. and Balmoral St. Parking is available on the surrounding streets and in the lots on nearby streets. Look for signage once you're at the building, or ask a security guard. ****************************************************************** For more information about MUUG, and its monthly meetings, check out their web server: https://muug.ca/ Help us promote this month's meeting, by putting this poster up on your workplace bulletin board or other suitable public message board: https://muug.ca/meetings/MUUGmeeting.pdf -- Manitoba UNIX User Group E-mail: <gedetil(a)muug.ca> c/o Gilbert E. Detillieux Web: http://muug.ca/ University of Manitoba Phone: (204)474-8161 Winnipeg MB CANADA R3T 2N2 Fax: (204)474-7609

1 0

new ECC computer options
by Trevor Cordes 04 Mar '17

04 Mar '17

I'm looking for input from anyone in the club who may be knowledgable, eyeballing, or has purchased a "workstation" system recently with ECC memory. (Note, ECC is imperative!) I don't need dual-socket, crazy speed, just something like a mid-range modern i7 with modern RAM, slots, etc. (i3 speeds won't cut it.) By "workstation" I mean it has to have at least one PCIe x16 (full 16 electrical) slot so it can have a good video card installed (and handle upgrades in the future). So some "servers" will qualify, although most do not have a true x16 slot (they'll say x16 but the small print will say x8 or x4 electrical), especially on the low end. I also need tower form factor, as rackmount will (unless 4U) be a pain to fit cards in, etc. Lastly, I really want a DIY setup where I buy the board, cpu, etc individually, though if a perfect premade system (i.e. Lenovo) exists that fits the bill I might consider it (must use no propreitary form factor parts, including mobo & PS). It's basically impossible since the discontinuation of the Intel D975 chipsets (nearly 10 years ago) to do Intel with ECC without buying a Xeon CPU. So I'm probably stuck buying a Xeon CPU and getting slower than i7 for double the price. Ugh. I'd really love to hear from the AMD nuts out there if there are any good AMD options, especially ones that are more enthusiast/desktop oriented rather than same-price-as-Xeon competitors. If I'm going to spend big bucks, I'll buy Intel: AMD needs to offer a compelling price advantage to pull me over. I still believe AMD could carve a niche for itself by offering desktop enthusiast chips / mobos with ECC at desktop, rather than server, prices. Maybe add $50 each to a mobo and cpu as the price premium. Not the 100% premium Intel wants just to get ECC. (I miss the old days when getting ECC was cheap/easy.) If AMD already has such a thing (I'm hoping!), great! I know I'm not the only one in the club interested in these answers. P.S. being a computer reseller, I don't need prices or store suggestions, I can just buy it all wholesale :-) It's the "this chipset plus that CPU" that I'm really looking for. Thanks!

3 11