jaclaz

Still NOTHING like a possible answer, but it is also possible that an actual release of the board software may be (part of) the problem. The CODEBOX parsing has a length limit, on 911CD I often need to split (very long) .inf iles into several different CODEBOXes, otherwise it won't simply work. As well some years ago there was a problem with an unfortunately common enough character sequence, which I cannot remember right now, but that made posted .inf's non-valid. Can you try re-posting here the particular contents of CODEBOX that gave you problems? jaclaz

VERY GOOD find! It's what I have (failing miserably ) attempted to ask help with since a couple of years: http://www.boot-land.net/forums/?showtopic=1507&st=1 (for "normal" 2K/XP/2003) Maybe this can be somehow "retro-fitted"..... jaclaz

NOT what you asked, but just for the record: CODE and CODEBOX "boxes" tend to behave strangely with different browsers/different versions of browsers. I remember an not so old previous version of Opera that did exactly what you describe. OT, but related, there are known problems with copy and paste (different results using different browsers): http://www.msfn.org/board/index.php?showto...24539&st=33 http://www.boot-land.net/forums/index.php?...=6492&st=10 It seems like it is a problem with the way data is "saved" to the clipboard: http://www.boot-land.net/forums/index.php?...=6492&st=15 jaclaz

NO. You need to chainload the loader, not the directory containing it. i.e. this would work on a CD: title Boot LiveXP on CD find --set-root /I386/SETUPLDR.BIN chainloader /I386/SETUPLDR.BIN Please be aware that CaSe Of FiLeNaMeS is important, always try using command line and enter commands one by one using the [TAB] autocompletion of grub4dos: http://diddy.boot-land.net/grub4dos/files/cli.htm to make sure that grub4dos "sees" the file in a given CaSe LeTtErInG. But the above WON'T work if the /I386 structure is on USB stick (or other HD-like media) as the hardcoded path in this case is /minint. (and usually filenames are lowercase on HD devices), thus it will become: title Boot LiveXP on CD find --set-root /minint/setupldr.bin chainloader /minin/setupldr.bin There are workarounds, but you should do a step at tthe time. Please note that all the above info, as well as a number of answers to the question and doubts you expressed in your previous post, is widely available on either boot-land or 911CD board. Particularly, since you seem like liking grub4dos and already started a rather ambitious project based on it, you should take some time browsing the entire forum: http://www.boot-land.net/forums/index.php?showforum=66 Doing a multiboot with several differnt OS's is not something you can learn and succeed with in a couple of days, you need to get familiar with a number of concepts, which right now you appear being not (yet ) fully familiar with. I do like your "experimental" approach, and I do like people attempting to re-invent the wheel : http://www.911cd.net/forums//index.php?&showtopic=22554 but within limits.... You appear to me as "falling" in the typical "newbie" (NO offence whatsoever intended ) mistake: trying to have everything at once. If you do not mind, let's try to make some "order" in the flux of your thoughts by enumerating some "basic" (and also a few "not-so-basic") advices and facts. General advice: Always start with one, and just one tutorial and objective in mind (not two, not three, just one). Always follow at first attempt a reported to be working method or tutorial or guide WITHOUT introducing ANY variation, of ANY kind. Even if the result of the tutorial is not EXACTLY what you wish to do, at first attempt you better follow the method TO THE LETTER and "deviate" (temporarily) from your original goal. Once and only once you have succeeded in following that method and got to the expected result for that method, repeat it introducing, ONE (or very few) at the time, the variations you think necessary to reach YOUR goal. Specific OS advice: DOS has a number of restrictions about the way it can be booted, but also a few advantages (see below about image booting) NT based systems have far less Linux has practically none (IF you know how to configure it properly) Image booting. Quite obviously, the most "clean" way to organize a multiboot system, generally speaking and specifically for USB sticks, is to have an image of some kind for each OS as a "monolithic" file that you can boot from. i.e., in a perfect world, you could have a ROOT directory with: grub4dos menu.lst dos622.img win98.img pmagic.iso slax.iso win2k.img winxp.img livexp.iso .... .... etc., etc. Unfortunately this is possible only up to a limited extent. Basically: ANY DOS can be booted from an image SOME - particularly built - Win9x/Me's can be booted from an image SOME - particularly built - NT or PE based systems, with the use of specific files from Server 2003 SP 1 or R2 can be booted from an image - Only RAMDISK based methods are nown to work at the moment SOME - particularly built - Linux distro's can be booted froma an image, but unlike the above, with the right knowledge ANY Linux distro can be configured to boot from an image Another "clean" way, that you initially took, but from which you seem to have deviated, is to multipartition the boot device, installing each different OS to a different partition. Of course the two approaches can be combined, just as an example you found the hard way how two Linux distro's based on the same "core" do not "like" being together. Now the disclaimer: though I could probably be described: http://en.wikipedia.org/wiki/Towel_Day as B) this only applies when we are talking about the initial phases of booting and about DOS and NT based systems, with the notable, but irrelevant as I see it, exception I make for Vista . When it comes to Linux, I know next to nothing of the internal ways and how to configure them. This said some (hopefully ) answers: Yep, see above. You need some time to learn and plan ahead. There are three ways: hardcoded path find --set-root method uuid method This: could become: You can use a "specific" tag file, to make sure that grub4dos roots to the right place, as an example place an empty file "this_is_slax609_on_usb.tag" in (hd0,1)/slax609/ then have: Or use a tag file to tag a partition, like putting an empty file "this_is_hd_0_1.tag" in (hd0,1)/ then have: For usage of uuid, check the grub4dos_readme.txt. By the way, always use - unless told not to do so - LATEST available grub4dos build, that you can find here: http://nufans.net/grub4dos/ older versions are here: http://nufans.net/grub4dos/history/ Actually this is the particular "field" on which the greatest amount of info is available, or at least the one where the most info is avialable in the grub4dos forum. About Gparted, your best option at the moment is to forget about it, and get partedmagic, which is already build in a way that it is .iso chainloadable. Read these: http://www.boot-land.net/forums/index.php?showtopic=5041 http://www.boot-land.net/forums/index.php?showtopic=4902 http://www.boot-land.net/forums/index.php?showtopic=6860 http://www.boot-land.net/forums/index.php?showtopic=5607 Take your time, and take it easy. jaclaz

Let's start form the beginning. Do you have a USB stick actually booting? (it doesn't matter which OS it is booting right now, only that it is booting) If not, the problem you are having is how to make that stick bootable. Once solved that, you can start thinking about adding to it grub4dos and multibooting. Which tells me nothing, unless you post some details. jaclaz

Yes, it may be the right thing, as long as you are seeking for the "easy" way. (other optins are slightly more complex) Just above your thread in the Forum: http://www.msfn.org/board/index.php?showforum=157 There are a number of "stickies" or pinned topics, second from top is WinsetupfromUSB thread: http://www.msfn.org/board/index.php?showtopic=120444 Yep, that's life, you have to take risks (or use money to buy a "safe" external DVD player) . jaclaz

Well, the (very old) thread you linked to is about grub4dos, not GRUB. I guess that you can find what you want by simply reading diddy's guide: http://www.boot-land.net/forums/index.php?showtopic=5187 Particularly, the guide presents 9 (nine) different ways to install/start grub4dos, of which I advise you to use methods #3 or #4, if your boot partition (as it should) already contains a NTLDR/NTDETECT.COM/BOOT.INI, method #5 if your "main" start is Vista. Also checking the numerous similar threads in the grub4dos forum: http://www.boot-land.net/forums/index.php?showforum=66 might help you. If you have any specific question, feel free to ask. jaclaz

Sure, life is tough. I am not sure what you mean by "arrow driven text menu with radio buttons" Grub4dos has limited "graphical capabilities", though latest versions have gfxmenu enabled, here is screenshot (oldish version, without gfxmenu): And here one with gfxmenu: You can use arrows to select an entry and also program hot-keys. If you want something really nice looking (depending on your graphical capabilities, of course ) you may want to have a look at Aerostudio, which can use grub4dos "under the hood": http://www.boot-land.net/forums/index.php?showforum=72 http://aerostudio.boot-land.net/ Here you can find a few screenshots on how nice it can be looking: http://www.damasgate.com/vb/810499-post31.html jaclaz

NO. As suggested in several places in this and the other threads, some cables for cellular phones integrate a TTL adapter, so that you could go that way, they should be easier to find. jaclaz

Do you need a verdict? I cannot say if I can give one , AFAIK it works nicely, of course it is NOT 100% compatible with "real" NT cmd.exe, but it is mostly. But there are workarounds possible for most of the "differences": http://groups.google.com/groups?hl=en&...&q=win95cmd Example, nesting FOR loops: http://groups.google.com/group/alt.msdos.b...&q=win95cmd In other words, do not expect to run "ANY" NT .cmd with win95cmd.exe, but most will do, and those that do not can be relatively easily fixed, whilst when using "standard" command.com you simply miss alltogether some features (SET /A, just to make an example). jaclaz P.S.:@Kelsenellenelvian Sorry for the OT/thread hijacking

OT, but FYI : http://www.boot-land.net/forums/?showtopic=2392 http://www.msfn.org/board/index.php?showtopic=100861 jaclaz

Have a LONGER look at grub4dos. One of it's features (missing in legacy GRUB and in Syslinux - but partially present in Siblo) is the ability to DIRECTLY chainboot OS system files, besides chainloading bootsectors: http://diddy.boot-land.net/grub4dos/files/boot.htm This is a perfectly valid menu entry in grub4dos: See?, no hands, no bootsectors.... B) You can also rename SETUPLDR.BIN to whatever you prefer, etc., etc. jaclaz

The link to flyakite guide it's not working, it's this one: http://flyakite.msfn.org/ If I were you, I would use a bootmanager, like grub4dos, on the AIO and leave bootsectors alone. Also, you should check the "Install XP from USB" section, there are a few things that may be of help for what you plan: http://www.msfn.org/board/index.php?showforum=157 It may take some time to go through it... Out of curiosity, any particualr reason for choosing Ruby? jaclaz

Actually, there is nothing you should be sorry about your English, but much about your aggressive attitude towards me, and other people participating to this thread. You may want, besides taking a crash course in manners and netiquette, review the Rules of this board: http://www.msfn.org/board/index.php?showtopic=18408 with particular attention to Rule #7.b jaclaz

Look, it seems to me like you have some problems with English, beside obviously lacking any form of politeness: the above means in plain English, I don't care who did it before, I am old enough to stay away from flamewars, you posted a link, I read it and let you know that what you extrapolated from it was not the whole story. jaclaz

Not that it means anything, but for the record ACE: http://www.acelaboratory.com/news.php announced a solution on 18th November: I presume that at the time of the announcement they alredy had made some tests, and had the sequence of commands at hand... jaclaz

Do a simple loopback test, first on the TTL TX/RX, then on the RS232 TX/RX. Details are here for the RS232: http://zone.ni.com/devzone/cda/tut/p/id/3450 http://zone.ni.com/devzone/cda/tut/p/id/3450#toc2 jaclaz

Well, your solution for #1 is not an answer to the question, it's the solution to the problem, a different thing. Is not possible to calculate the chances of a drive x being part of the affected lot as we do not know exact volumes of production, nor ratio of "problematic tester/good tester" in the actual factories. If we knew this, we would have some of the factors involved, example (assumptions): Seagate produced 5,000,000 drives of the family in 2008. On average 5,000,000/360=÷13,000 drives/day (in a 24/7 timetable, which cannot say if it's how factories work) Let's say that these drives are manufactured in 3 factories, thus each factory outputs roughly 4,333 drives/day, i.e. 180 drives/hour How long does it take to produce/assemble a single drive? Let's say that 10 minutes are needed, thus the parallelism of the production line has to be 3, i.e. they need to have 30 identical lines in the factory, each one will produce 6 drives per hour, 30x6=180 . If just one of the testers is of the faulty type, 1/30 of drives will be affected, i.e. 3.33% If half an hour (30 minutes) is needed, parallelism becomes 90, 2*90=180 and chances become 1/90=1.11% But without actual data, all the above cannot represent "real world", though the initially assumed 0.002 would mean, an overall parallelism of 1/0.002=500, which you may imagine as 3 factories having 166 production lines or 10 factories with 50 lines each.....in which just one single tester is of the defective type, a bit hard to believe (or not?) As well, your solution for #2 is not an answer to the question, the chances that an affected drive will develop the problem are 100% in enough time. This is due to the nature of the problem, a circular buffer will hit one of the (320+n*256) values, sooner or later. Yep, that is the most "questionable" data, can mean ANYTHING bigger than 0 and smaller than 1. (i.e. 0.000%<p<100.000%) jaclaz

Let's start again. If it's shown as BUSY, it is damaged in it's firmware. There is NO data recovery program (software) on earth capable to recover from a BUSY (firmware) problem, it simply cannot see the drive. You cannot possibly have applied the new firmware to a locked drive, you need a RS232-TTL interface and unlock the drive: http://www.msfn.org/board/index.php?showtopic=128807 or send it to Seagate. jaclaz

Which is a direct download link for the app by ilko_t, ALREADY referenced, here: http://www.msfn.org/board/index.php?showtopic=120444 jaclaz

Yes , see cdob's report here: http://www.msfn.org/board/index.php?showto...20&start=20 He even succeeded in booting XP from exfat32 using the "XP Kansas City Shuffle"! The only file actually needed is the exfat.sys, to access Read/Write: http://www.911cd.net/forums//index.php?sho...c=22584&hl= all the others are for formatting and a few other non-basic functions. It's still a "mistery" the modified shell32.dll. jaclaz

http://consumerist.com/5142062/seagate-cen...-drive-failures jaclaz

TRADUZIONE ITALIANA della guida di Gradius2, l'originale (con immagini e links) è qui: http://www.msfn.org/board/index.php?showtopic=128807 (la traduzione si basa sulla versione online il 28/01/2009) La traduzione è stata fatta con la massima accuratezza di cui sono capace, ed il più fedelmentte possibile, ho solo eliminato alcuni punti più tecnici (che chi è in grado può agevolmente ricavare dall'originale e qualche commento/link non essenziali. NESSUNA responsabilità per errori ed omissioni. La Soluzione per i problemi degli hard disks Seagate 7200.11 (versione finale rivista e corretta) Attenzione: Questa guida può NON essere alla portata di tutti. Non è così difficile ASSASSINARE il vostro disco rigido se non seguite ALLA LETTERA le istruzioni, agendo con la massima cautela. NON sono responsabile per qualsiasi danno o perdita di dati che possa accadere (sebbene non vi siano stati finora casi, un incidente può sempre succedere). Possono avvenire comunemente due errori sui dischi Seagate 7200.11 (causati da bugs nel firmware) - 0 LBA error: quando il BIOS riconosce il drive all'avvio, ma come un drive di dimensione 0 Mb. - BSY error: in un particolare stato il disco rigido può entrare in uno stato di stop, o BuSY (occupato). In questo caso il disco non viene riconosciuto dal BIOS all'avvio. La spiegazione di Seagate: Avrete bisogno di: - 1 (uno) interfaccia RS232-TTL - 1 (uno) cavo RS232 cable SOLO in caso di un PC senza porta seriale, (RS232) 1 (uno) cavo convertitore USB->Seriale Anche alcuni cavi USB di interfaccia specifica per cellulari possono essere adattati. - 1 (uno) kit di spezzoni di cavo con connettori o anche con "coccodrilli" - 1 (uno) cacciavite mini Torx 6 - non sono facilissimi da trovare - generalmente le serie si fermano al torx 7 - spesso il 6 si trova nelle serie che hanno i venditori ambulanti Il resto: - 1 (uno) computer con un software di trasmissione testo su seriale/Telnet (come HyperTerminal o similare) ed una porta seriale RS232 o USB. - Una versione aggiornata del firmware Seagate (Released in 22/01/2009 and 24/01/2009 per i dischi che potrebbero essere affetti). http://seagate.custkb.com/seagate/crm/self...sp?DocId=207931 - Un po' di pazienza e pratica. Connessioni TX (trasmissione) e RX (Ricezione) su un SATA Seagate 3.5": Foto del disco: Connettete le porte TX e RX all'adattatore TTL, usando i cavetti con terminali o coccodrilli, o comunque con una connessione efficace. Ricordatevi che le porte TX e RX sono direzionali, quindio dovranno essere connesse così: - Da PC (TX) a >>> RX (sul disco Seagate) - Da PC (RX) a <<< TX (sul disco Seagate) Connettete l'alimentazione, lo 0 (GND) e i +5V all'alimentatore del PC (cavi Nero=GND Rosso=+5V) od altra fonte. (NON usate il Giallo=+12V) Per stare sul sicuro, è consigliabile usare come alimentazione una batteria da 3v tipo CR2032 o due batterie stilo o ministilo messe in serie. Un idea di come fare (io ho usato un cavo USB-> RS232): Avviate HyperTerminal o il programma che usate (con XP: START->Tutti i Programmi->Accessori->Comunicazioni->Hyperterminal) Configurate la porta RS232 con: 38400 8 N 1 (come qui): Riparare il problema "0 LBA": Nota: Tutti i comandi sono CaSe SenSiTivE, usate le minuscole e MAIUSCOLE esattamente come nelle istruzioni! E' necessaria una formattazione rapida per rigenerare le partizioni, com m0,2,2 e 0 e virgole concluse con un numero "validkey" oppure 22. ATTENZIONE: I comandi DEVONO essere usati ESATTAMENTE come mostrato, un errore, anche minimo può rendere il drive inutilizzabile o creare una perdita di dati! Controllate, ricontrollate, ed infine controllate ancora un volta quanto immesso PRIMA di premere Invio! Il comando da usare è: F3 T>m0,2,2,0,0,0,0,22 e premere [invio] Adesso aspettate circa 15÷30 secondi fino a che apparirà qualcosa del genere (può variare da drive a drive): SOLO DOPO che questi messaggi sono apparsi staccare l'alimentazione. Congratulazioni, ce l'avete fatta! (Addio danaro! Sarei potuto diventare ricco!) Se credete, una donazione (a Gradius2) è bene accetta. Riparare il problema "BSY o CC ovvero BUSY": (si raccomanda di fare pratica SENZA alimentazione prima di provare questo metodo!) Nota: Tutti i comandi sono CaSe SenSiTivE, usate le minuscole e MAIUSCOLE esattamente come nelle istruzioni! Prima cosa, assicuratevi di avere un piano di lavoro comodo, bene illuminato e sgombro. Secondo, dovrte smontare dal disco il circuito stampato, svitando le viti con il cacciavite Torx 6. Terzo, connettete i cavetti TX e RX del convertitore RS232, controllate che siano connessi correttamente e attacate il connettore di alimentazione alla scheda (dall'alimentatore del PC) Spiegazione: Il problema è che il disco è bloccato in uno stato di occupato "BUSY" che blocca le comunicazioni attraverso la porta SATA, per cui il BIOS non riesce a trovare il disco all'avvio. Nel terminale vedrete qualcosa tipo (LED: 000000CC FAddr: 0024A051). In questo caso la sessione del terminale è bloccata e nessun comando viene accettato. Per aggirare il problema è necessario staccare il circuito stampato dal disco, in modo che il messaggio LED: 000000CC FAddr: 0024A051 non possa essere inviato nella sessione al terminale Telnet. Dopo aver connesso l'alimentazione al circuito stampato (staccato dal disco), premete Ctrl+Z (anche più volte) in Hyperterminal, dovreste visualizzare questo prompt: Se non lo vedete, ri-controllate i collegamenti TX/RX dell'interfaccia RS-232, potreste avere i fili invertiti. Adesso è necessario accedere al livello 2, immettete: F3 T>/2 [Invio] Il prompt diventerà: Adesso immettete (per fermare il motore): F3 2>Z [Invio] Dovreste vedere il messaggio: Ora, la parte DIFFICILE. Provate, provate, provate senza alimentazione prima. Dovete essere assolutamente sicuri che le viti non tocchino il circuito stampato dove non devono, un errore, anche uno solo, e la scheda è andata PER SEMPRE. Se siete già sicuri ed avete fatto le prove, potete cominciare. Dopo aver fatto la procedura di cui sopra, e con l'alimentazione ancora attaccata, così come i cavetti TX/RX, rimontate la scheda sul disco FACENDO ATTENZIONE di non mettere nulla a massa, che NESSUN cavo si disconnetta, etc.. E' essenziale che il rimontaggio avvenga dopo avere immesso i comandi di cui sopra SENZA interrompere l'alimentazione e la sessione Telnet aperta. Un buon sistema è quello (ad alimentazione spenta) di staccare la scheda, e rimetterla al suo posto mettendo una striscia di isolante (cartoncino, NON plastica che potrebbe avere una carica statica) tra scheda e disco, quindi riabboccare appena le viti, lasciandole abbastanza lente da poter sfilare la striscia isolante. Quando la prima parte della procedura ha avuto successo, basta semplicemente sfilare il cartoncino e premere la scheda al suo posto (stringendo un pò le viti). Adesso immettete (per avviare il motore): F3 2>U [Invio] Dovreste vedere: Andate al Livello 1 (immettendo /1): F3 2>/1 [Invio] Cancellazione S.M.A.R.T. erase (create settore S.M.A.R.T.): F3 1>N1 [Invio] Cancellazione G-List (certificate il cilindro riservato): F3 1>/T [Invio] F3 T>i4,1,22 [Invio] Spegnete il disco (staccate il cavo di alimentazione) MOLTO IMPORTANTE! Attendete almeno 10 secondi e riaccendetelo (riattaccate il connettore) In Hyperterminal premete CTRL+Z ed immettete: Partition regeneration: F3 T>m0,2,2,,,,,22 [Invio] Adesso aspettate circa 15÷30 secondi fino a che apparirà qualcosa del genere (può variare da drive a drive): Se questo messaggio non appare, qualcosa non ha funzionato, NON STACCATE l'alimentazione, il disco può diventare inutilizzabile, STATE ATTENTI! SOLO DOPO un messaggio di questo tipo potete spegnere il tutto. Congratulazioni, ce l'avete fatta! (Addio danaro! Sarei potuto diventare ricco!) Se credete, una donazione (a Gradius2) è bene accetta. Qui c'è una guida con foto: http://www.msfn.org/board/index.php?showto...8807&st=100 **************************** N O T E **************************** Tenete presente che io (jaclaz): non ho un hard disk di quelli che presentano il problema non ho mai eseguito personalmente la procedura non so praticamente nulla del "linguaggio" usato da Seagate Se avete bisogno di aiuto, per favore postate qui, io proverò a tradurre la domanda e girarla in inglese nel thread "principale" traducendo un'eventiuale risposta dall'inglese. Se qualcosa non vi è PERFETTAMENTE chiaro, se non siete ASSOLUTAMENTE CERTI di essere in grado di seguire questa guida: chiedete aiuto, chiarimenti PRIMA di fare qualcosa di ireeparabile considerate l'idea di contattare Seagate, ed una volta avuta assicurazione che ripareranno il disco in garanzia E che lo renderanno con tutti i dati, mandatelo a loro Il mio consiglio personale, non appena siete riusciti a far vedere il disco dal BIOS, ed avete i dati nuovamente accessibili, è di copiare tutti i dati su un altro disco PRIMA di applicare la nuova versione corretta del firmware. (Nota bene, questo è quello che avreste dovuto fare comunque, fate sempre ALMENO una copia di riserva dei dati, MEGLIO due.) Se anche non potete permettervi una piccola donazione a Gradius2, anche un semplice nel thread principale: http://www.msfn.org/board/index.php?showtopic=128807 sarà gradito. jaclaz

Well, using the "safety factor", I already used a speculative "base" of 10,000,000. As already posted, them being 5,000,000 instead: And in any case, the "objective" of the speculation: was NOT to determine exactly: how many drives were produced how many drives are affected and have developed the problem only to check if it should be a matter of a few hundreds (a "few" or a "handful") as opposed to several thousands (a "lot" or "too many to count"). jaclaz

Additionally, and JFYI, since there are a number of reports of memdisk failing with UBCD on some hardware, there is an ongoing project to add grub4dos booting to UBCD (without using memdisk). If everything goes well, UBCD 5.00 beta11 or beta 12 should be "grub4dos aware". http://www.boot-land.net/forums/index.php?showtopic=6119 jaclaz