|
Technical News and Views from a hole in the ground: 2008 TimeLine 2008: page edited: Friday, July 04, 2008 Time Line Base Separate page about the Fujitsu, eBay, GreatDeals00 charade |
|||||||||
| January 1st., 2008: Searching for 21" CRT
monitors. Three are needed, and one is a replacement for the
mis-coloured Hewlett Packard Ultra VGA 1600: long name for a faulty
box. No real hurry,
because I don't know when the Dell 650 Precision is to be removed
from Computer Supply House. This is still the holiday season, and I
need to spend on other people, not just on myself. What, selfishness
is lacking,
eh? The problems with the local network domain seem to be at an end: only one computer has need of the ASR operation. That is the white box, Oxford, that won't accept floppies into its A: drive. I will have to determine what's wrong there. Firstly, however, I need to install casters on the primitive box that it and the ML570 stand upon, and with at least 250lb of books contained on the shelves. That will help me work with the computers, because the ML570 has serious weight. It's not easy turning it on its side to open the case but it's much easier to work with the constituent parts when it's in that position. With space constraints in this room demanding alleviation, then a movable "table" will certainly help, but the casters need to be heavy duty. I shall go down to Preston Hardware: their stock is excellent, and I can trust them for quality. The ML570 G Server, now the sole DC in the network, is to have a new adapter. Having gained an U320 LSI adapter on eBay, which is in transit, I am trying, from the service manual pdf, to find out if the cables will need to be replaced to suit the requirements. There is little that is mentioned about the yellow and blue SCSI cables that attach to the drive cages. One thing that I must do is back up all of the drives installed in the cages, whether RAID arrays or JBODs. I don't know whether the adapter will import designations made under the currently installed Adaptec 3200s controller. That will be removed and kept as a spare until needed elsewhere: the network slowly grows, and RAID arrays are needed for reliable storage. Thursday, January 3rd., 2008: I realised that the A: drive connector on Oxford was probably reversed. That was easily fixed after I took under an hour to put casters on the wooden support box and replace Uncle Tom Cobley. Moving the book/computer 'stand' is now easier than it was, although there is serious weight involved. However, XP Pro on the white box, Oxford, gave me page file errors and refused to start. Windows almost started and immediately closed down to reboot without allowing one to capture the screen or have time to read it. Shutting down and swapping all of the memory sticks cured that, and now I am performing an ASR on Oxford whilst I write this. Microsoft requested a report on a major start up fault. What it offered as a reason was inapplicable, so I told them what had happened and how I fixed it. Why not? Some people refuse to advise Microsoft, but then wonder why they can't find a cure later. Friday, January 4th., 2008: GreatDeals00 indeed. The last two zeros imply something. Yet another 147GB U320 drive is giving me problems. I have bought too many drives to be fooled by thinking that the failure rate of this vendor's goods is normal. Four of eighteen of this capacity have now failed. Given my experiences with other drives, of all types, whatever this dealer is selling are fake goods. Here is the email I sent to GreatDeals00
today:
Best Deals Anywhere
welcomes you our ebay store. We offer low prices with great
quality products everyday. All auctions start at .01 and we have
new products coming in daily. We are one of Ebays top sellers so
you can bid with confidence each and every time. We have served
over 200,000 customers and counting. +++++++++++++++++++++++++++++++++++++++ Here is my statement The above is taken from your ungrammatical blurb contained in an offer for a Fujitsu 147GB U320 drive selling today.
Yet another of the
eighteen drives of this capacity that I have bought from your
store has now failed. This makes five when one includes the
replacement drive that never worked.
It is clear from the
amount of goods sold by you that people are being scammed. That
the feedback states how happy they are is ridiculous: they can
have no idea what they are buying. I had, late last year, approached eBay: their statement that anything they do must be kept private, given that this vendor's drives constitute a significant proportion of all of the U320 drives sold within their structure, means that the money they make from him clouds their thinking. The problem is serious: I cannot afford to have drives fail like this, for any reason. Here is the email sent to me by eBay after my complaint:
Dear eBay member,
Thank you for your report. We will thoroughly review the listings you have reported for violations of eBay policy. We sincerely appreciate your efforts to report such listings or posts to us. Please keep in mind that account information is confidential. Therefore, whatever action we may take will be between eBay and the seller. We will not be able to keep you abreast of any developments.(my emphasis) If we determine that the listing does violate policy, we may choose to: - send the seller an informational alert; - remove the listing or post; or - suspend the seller. Account suspensions are usually reserved for those sellers that continuously disregard policy. If you would like to review other Listing Policies, please visit: http://pages.ebay.com/help/policies/listing-ov.html To report any other listing violation, please visit the link below: http://pages.ebay.com/help/contact_us/_base/index_selection.html We appreciate your concerns and we thank you for helping to keep eBay a safe and reputable forum in which to conduct business. Sincerely, eBay Customer Support _____________________________________________ Whether you're new to eBay or an experienced buyer and seller, the eBay Security & Resolution Center can help you protect yourself on eBay and online. For more information, please click the "Security Center" link at the bottom of most eBay pages. _____________________________________________ For our latest announcements, please check: http://www2.ebay.com/aw/announce.shtml ++++++++++++++++++++++++++++++++++++++++++++++++++++ End of email I am not happy about this Star Council attitude for customers; where can one determine that eBay actually looks at one's complaints? If one goes to the New York Times, please search for eBay problems in the blogs thereon. Illumination will appear. Monday, January 7th., 2008: I thought that another of the Jas man's drives had quit. Luckily not, because I opened the Dell case and checked all of the connections. That cured the non-appearance of the drive under the LSI controller. There were some chkdsk errors when the drive was automatically checked on start up. These were cured, and although FireFox needed to be reinstalled, it found the previous settings and started without fault. I need to contact the authorities in California regarding Great Deals (Best Deals Anywhere), but that will have to wait until I have time: father is unwell. Tuesday, January 8th., 2008: I was woken in the night by the sound of an alarm. It was the ML570, and the source was the Adaptec 3200s adapter. On it was a Jas man 147GB U320 that had failed, and I was being told about it. Not another one: so, I arose and for a couple of hours worked at replacing the drive and starting on the build of a new array. This afternoon, returning to the cell, I phoned Fujitsu USA, and spoke to a techie in hard drive support. He was interested enough in my tale of woe to give me his email address with the request to send him a report about my experiences with GreatDeals00, Best Deals Anywhere, ha! This I shall do tonight or tomorrow morning. Maybe this is the beginning of problems for eBay, too. They, in my mind, are in collusion. Thursday, January 10th.,
2008: Awoken at 06:00 hours by the Adaptec 3200s: yet
another of the Jas man's U320 147GB drives has failed. This was
probably why the array was taking such a long time to build.
There was only 2% completed after two days, and when I checked visually
last night the faulty drive that gave notice this morning had
its drive action light on constantly. Should have reasoned why,
shouldn't I? Tuesday, January 15th., 2008: Problem with FireFox on one XP Pro computer: no matter what, nothing when asked for is saved in options. Nothing however as bad as IE on W2003 Server boxes. Saturday, January 19th., 2008: I had received a couple of drives, to replace others in Devon, an IBM Windows 2003 Server. Automated System Restore was used. The result? Of the drives that were not to be replaced all were totally scrambled, and also unreadable in any other computer that was used to check them. One drive on the chain, that was chosen by ASR to be where the OS would install itself, remained usable, but not a single one of the others. Two of the SCSI drives, of the three in the hot-swap cage, had to be changed to basic before they could be used: hence all data was lost. I was able to change them back to dynamic later. As well, ASR under Windows 2003 constantly tried to find the lost back up files. It would not allow me to make a clean install. I need the real Winternals tools in a hurry so I could cure this nonsense. Given the situation with Windows 2003 Enterprise I decided to go with XP Pro SP2, and the machine is running much more quickly. One reason for that was because, having removed the IDE drives, I replaced the eServer 220 box's storage innards with all SCSI objects. All of the data and programmes had been lost, so that is going to be a long process to bring up to scratch. It's a good job that I have other computers that can take the workload in the meantime. One thing that I did like was that the machine joined the network very easily and I was able to import the users I wanted with no bother whatsoever. Neat that that happened, even if the cause of OS change was an ASR disaster. Monday, January 21st., 2008: Several requests to re-write the pages referencing Fujitsu drives came in. I have, because others are suffering the same problems that arose within my systems, decided that it is only fair, simply because I was repaid, to ensure that facts are available for other complainants. Freedom of speech, methinks. After having the snafu with ASR on Devon, one of the Win 2003 servers, which made the SCSI drives unreadable, I had installed an IDE CD drive whilst trying to fix the computer. That has been removed and replaced with another SCSI hard drive. The two CD-RW external SCSI drives will work with whatever I need to do with them. Now, that is fine because I created the six floppy boot disks that allows one to start the XP Pro SP2 setup without having an IDE CD drive. Given that the IBM BIOS on the eServer can't recognise a SCSI CD one is not allowed to install Windows Server 2003 or higher. There are no floppy options for Windows Server versions 2003 or 2008, which makes things difficult. I need to buy two more drives and that will finalise drive purchases for a while. Assuming no more of the U320 ModusLnk drives given problems. Once they have been running for a while, one can assume them to continue to do so. Sod's Law. The usual problem with FrontPage arose because I had to download all of my sites to a different server for storage purposes, given that Devon had been the originating storage box. Once that had gone to the ASR hell, I had no option but to download to Rutland, the ML570. However, FrontPage believes that all of the files have had a change in authority, thence need replacing. That takes hours. There is no remedy. Thursday, January 24th., 2008: Bits and pieces obtained today, to accommodate the drives that arrived. Two more U320 147GB (not from GreatDeals00) 15k items to go with the three U320 73GB 15k on the native LSI bus in the second Dell Precision 650 that is now silently sitting next to its twin. I shall purchase PC3200 SDRAM for it in February. There are also two 1.6" 73GB U160 10k monsters, and a spare, to go in a ProLiant 800 that I shall bring back from storage. I went down there today for two objectives: the first was to photograph the Haier air conditioner so that the manufacturer is made aware that it is a physical entity without any labels. I spoke with the web based HomeDepot support personnel yesterday, and they managed to persuade Haier to give me a replacement. That will done when I email them scanned images of the photographs and send them, yet again, the documentation that I received via email in the first place. Plus the scanned delivery notice. That's all on the web page that I produced about this fubar. What a joke, to have to go through all of this, simply because they cannot accept that faulty equipment leaves their factory. Heavy work, moving boxes filled with books around. I have also asked the manager at the storage site for a larger room. Mine is almost full, and, since I found another dead mouse in a trap laid there, it is important to take everything out to check for damage by tiny teeth. The second objective was to see if the hot-swap drive cage would indeed fit into the ProLiant 800. Yes, indeed! That will result in a machine with spare slots to use for repairing drives. The two 73GB drives will fit just below the CD-RW and run XP Pro SP2. I have another six slotted cage, too, so that later on I might involve another ProLiant 800 in the repair scene. Using Gibson's, or another's, software takes a long time to complete its job. One thing about these SCSI drives that I have discovered: some give errors depending on the adapter involved, and give no problems in other machines. That especially goes for one 73GB drive, U320 from GreatDeals00 that won't work with an Adaptec 29160n but will work with an Adaptec 3200s. Strange, what? Must be the drivers involved. The 3200s is more sophisticated given that it is a RAID card. Finally today, I installed one of the two U320 LSI PCI adapters I bought recently into Leicester, the second Dell Precision. This will be one really nice machine when the RAM is installed, and everything is spinning along, pushing four LCD monitors, two above two. That's what I'm aiming for, 3300 x 2100 total resolution. Light in weight, low in power consumption compared with the two 21" CRT monsters that the first Dell Precision, Monmouth, propels. That means five LCDs are required, because one can be placed in Rutland, where the poor old CRT is removing yellow from the screen. Monday, January 28th., 2008: Today was spent looking after father on a trip to the Civic. On the side, thought about problems with ASR on Rutland, the AD Windows 2003 Enterprise server. I think that I'll buy Acronis Migrate Easy 7.0, because that should allow me to copy the boot drive, which is an U320 73GB drive, to one that sits more easily alongside its U160 73GB neighbour on the Adaptec 29160n. That would free the U320 drive to fit into the spare slot on one of the arrays, that appeared when a Jas man U320 drive failed. It matters little to me whether its a 147 or a 73GB example. That can be changed at a later date. The idea is to have two of the 1.6" Seagate U160 drives together, which would allow both to run at expected speeds on a native bus. (But, see below.)
Tuesday, January 29th., 2008: The
reason why the ASR for Rutland failed: one drive fractionally
less in size stops the process. Both drives are nominally 73GB
in size.
Wednesday, January 30th., 2008: Well,
well, another snafu now with Acronis True Image. It, too,
will not work with dynamic drives on an XP Pro SP2+ computer.
What is the use of a programme that will not work with commonly
found hard drives? Does one have to spend a grand US on their
server software to do this? On the Acronis website sales pages
one discovers nothing mentioned about dynamic drives. On top of
that, on the page for True Image Echo Enterprise Server, an
extra cost of US$299 is needed to do a universal restore. This
does not make me happy. I have been thinking about opening a Support Ticket with Microsoft concerning the ASR option for Windows 2003 Enterprise server (or for XP as it happens). I want to replace a same-size hard drive. Just because the "new" drive is less than 1% smaller than the "old" one, ASR won't work, even if the originating "old" drive has less than 4GB in use, out of its nominal 68.4GB. Rubbish, what? Note in the two images immediately above that the error message is identical. No mention of this in any of their advertising. Going to their search pages one discovers this: Dynamic storage support in Acronis True Image Server. The first sentence on the page states: "Dynamic disks are a central part of Windows storage management on the Server and XP Professional platforms (my emphasis), but the same features that make them so powerful make them difficult for conventional disk image software and backups to handle". So what, I say. If they are a basic part, being themselves dynamic, why can't software for normal operations become normal??? The dynamic drives I wish to alter are not on arrays, they are single drives. That they might in future become part of another arrangement is irrelevant. What I need is simple, good software to meet simple, ordinarily met problems.
Thursday, January 31st., 2008: An answer,
a non-answer, from Acronis: Thank you. Acronis Customer Service Department As the two images above relate, I cannot use the programme. So, Boilerplate 101 here. I know damned well what dynamic drives are, but considering that the server products can handle it, at least they state they can under a New! heading, it would seem that there was a considerable demand by IT guys who needed that facility. Given how common dynamic drives are in the universe, I can understand that handling them is an important requirement. On a related subject, I have opened a support ticket with Microsoft about ASR. There is a KB page at Microsoft (http://support.microsoft.com/kb/314686) which states that differences in disk geometry of drives of nominal commensurate size can stop the operation. I talked to the support guys and remarked that this is a basic flaw in the programme. What, for instance, is wrong with the scenario that might arise should one only have a spare drive smaller than that currently installed and a crisis occurs. Perhaps a drive failure happens, and you need your AD machine back in a hurry. A new install of the OS and AD to follow is not a trivial operation. There are massive updates also implicated. If one only has a smaller drive one is at an impasse, with current ASR programming status. I understand that there is to be a replacement for ASR in the works, but this is what is available right now (not true, it was something to do with WSUS, instead). It's a situation that is exacerbated by the fact that Windows XP Pro knows how much free space there is on every drive. Given that, it should be easy to determine if the system/boot drive would fit its used space onto a smaller replacement. Given that the drive I want to use is fractionally smaller than the one currently installed, I can't see why it shouldn't be possible, with a decent algorithm. We'll know if there is a fix within a few days, I hope. Friday, February 1st.,
2008: The Acronis guy sent this, totally
inconsequential, reply: ASR:
after hours on the phone with
Microsoft, New Delhi, India, we were unable to fix the ASR
problem. So, today, I installed a 1.6" (half-height) U160 73GB drive in the place
of the U320, which is now in the bottom cage of the ML570. Now that Adaptec Storage Manager is reinstalled, it
would seem that the computer is now working as I had hoped. One
thing is that even with JRE installed, IE7 will not allow me to
log in to register and reach my MS TechNet login page. FireFox does, and that was
the browser that I used when working with the MS Techie in
India, Sanjay. Nice guy, very helpful, in as much as his crew could
do. Conclusively, ASR has this, to me, inexcusable failing,
which was explained earlier. Aaaargh! as they say. I have had
problems with SID errors; with trust relationships. Active
Directory is not working properly. The network
partially works, in that every other computer on the network
can connect with themselves and with the DC, Rutland, with the
newly installed U160 drive. But Rutland can not connect with
any member of the domain.
Forget this for a bunch of trouser
snakes: I reinstalled/reshuffled/removed drives so that the 1" 73GB U320
(the good, working boot/system drive) was returned and I placed a 147GB U320
alongside it, both connected to the Adaptec 29160N. Both drives work and having
restored the data to the 147GB drive, everything is seemingly back
in good working
order. Once more there is a gap in the ML570 lower cage. But,
never mind, I can fill it later on. To conclude: if only ASR
was dynamic, and total used space was the basic quantity
expected on a replacement. Wednesday, February
6th., 2008: The FP extensions had to be reinstalled on
this site today. Fixed by 20:00hrs, after a day of frustration.
At least I was able to photograph and email, as well as upload,
the Haier air conditioner replacement request. Then I was able
to photograph my cell and upload the images to another page on
this site. The snow here in Ottawa was sloppy, allied to the
effects of salt on road corners, and possessing Sorel rubber
boots was a
bonus walking out and around. Father is surviving, and seems a
little better that he was, so that I do hope that an operation
is deemed acceptable for someone of his age. That would make it more
likely that he would survive, until at least July when the family visit.
Sunday, February 10th., 2008:
I had the money, so I
bought and downloaded Ad-Aware. What a mistake. It will start
and _never_ finish. Useless. It stalls on folders and one can
leave it for hours and it will not complete its self-set tasks.
Monday, February 11th., 2008:
Ad-aware fubar.
Unresponsive programme that has been reported to MS. If I left
this alone the clock would keep running, but nothing else
happens.
eBay problem, again:
One new thing is that I received a box full of old Compaq
hot-sway caddies. The box was worn through from the inside,
because the vendor had not wrapped the items. I have never
received items unwrapped before. Shoddy idea of customer service:
I was lucky to receive the purchased total. The vendor blames me
for poor communications in not letting his company know before I
posted feedback. Sod that. Here are the emails that passed
between us. The first one was because he was trying to charge me
$20 post and packing per item, all of which fitted into one small box.
They eventually
came all jumbled together, in a worn cardboard container.
Dear rns-auctions,
Dear transubstantial,
Dear rns-auctions,
Dear transubstantial, This is hilarious, isn't it? One
believes that the seller has a moral obligation to help the
buyer? Well, the customer is clearly not always correct, what?
Monday, February 18th., 2008:
I managed to fix the QuickTime query lines that proliferated
throughout FireFox, of whatever version, on one computer, the
Dell Precision named Monmouth. There is a massive extant problem
and I have heard from others that they have never been able to
correct it.
However, I deleted the
requisite profile and the FireFox folder from Monmouth, and then
copied those from another computer, Oxford, the white box, which
has the same version, but no problems apparent. They are both
running Mozilla Firefox 3 Beta 2. They now both work properly,
whereas one was giving me fits not being able to open most
third-party pages.
Ad-aware 200x or whatever, and it is the latest version that I
am paying for. It is, once more, unresponsive. The clock will
run until the computer is recycled or otherwise disposed of. It
will not complete. Note that I went to the Civic with Dad, and
when I came back the programme was again just running the clock.
Note the Scan time:
Wednesday, February 20th.,
2008: A reply from Lavasoft, the Ad-Aware people. Why should
I bother? The reply included the request to update to the latest
version, which I had already done. There was a statement that
Ad-Aware could conflict with SpyBot. That is rot, since the
programmes are not run at the same time, and don't conflict on
any other machine. Note that the problem arises only on the
machine that has the registered, annual fee paid version.
Significant?
Saturday, March 29th., 2008:
Nothing much of late, except that, even though one of
my 2003 Enterprise servers seems to be working, I cannot install
any updates, plus the sharing seems all to cock. Probably, the
best thing would be a complete reinstallation of the OS. Not
trivial, and perhaps its better to also have an AD backup?
And, I have discovered a couple
of basic errors: the main one being that the slow signup was
caused by me having entered a false DNS server IP address for
the LAN. Dumb as a Red Thatch, moi. Then, having fixed that, I
was able to make Lancaster the backup DC, because now dcpromo
could find the domain controller. That's for safety anyway, in
case of failure on the main box, Rutland.
Also, having fixed an error with
BITS, I am now able to download all of the necessary updates
onto Lancaster. That makes me happier. Now to fix the reason why
Event Viewer is missing from the Administrative Tools listing,
although I can find it using Help.
Tuesday, April 2nd., 2008:
I had yesterday received a Compaq cage for my ML570, that allows
SCA drives in lieu of 68-pin for the two-seater in the bottom
left front of the server. I could have installed it yesterday
morning, but it was All Fools Day, wasn't it? In any case, I
have never had such an easy job, and the server booted up
cleanly, with the Adaptec 29160n accepting the signals from the
Compaq ProLiant 4L2I at ID 15 on the chain. This means that any
drive in the ML570 can now be removed by clicking on the caddy
rather than opening the case. This box is really heavy! Nice to
have something work so easily.
Periodically, one receives emails
from Network Solutions asking one to update one's registration
details. I changed one field to remove a non-existent fax
number, but could not edit any other field. The arcane method
that Network Solution uses is not user friendly, and they have
now allocated a lock on editing any details of my account with
them. Given that I pay for the service, that the servers
actually in use at SoftCom have been stable for some years, this
level of support from Network Solutions is utterly useless. I
believe that I should be able to edit my details at any time,
given that I, and no one else, am the owner of the domain names
they hold: mattoid.com and mattoid.net.
Sunday, April 20th., 2008:
I have worked out why I couldn't alter the Network Solution
fields: there are two passwords, and one of them allows every
field to be altered, and the other only allows one field. Huh?
Then arose another problem: a PnP
KVM will not work with my Compaq servers, and I can't find out
why, although the fact that I couldn't use floppies to install
Windows 2003 Server, and SmartStart on my ancient ProLiant only
allows those, then I can't use SmartStart in its proper fashion.
Clear as mud? Don't worry, I can't figure it out either.
Late today, delayed write failing
on F:\$Mft, and that I thought was odd. Rebooting removed the
drive, and lots of data. Closing down, checking all of the
connections, finding one on a hard drive that was maybe a hair's breadth
loose and
bang, it is back working. Just a small flaw, but the Dell
Precision was acting like a laptop.
Saturday, May 10th., 2008:
Nothing much of late, except that I am trying to tie the two DCs
to a KVM, with little luck. Also, the email server at NCF is
kaput. Two days now and no sending service. Not clever: where is
the reserve machine? Before I checked their website I had
thought it was an Outlook problem. Not so, and I'm glad of that,
because it's been stable for a long time.
Saturday, May 17th., 2008:
More than two days trying to start up the second Dell Precision
650 that I obtained from Computer Supply House, on O'Connor in
Ottawa. The problem is/was somewhere in the LSI MPT SCSI BIOS,
which for a considerable time would not complete. Finally, it
will start but only in narrow configuration. I have posted on
the Petri site to see if anyone has an answer. LSI support pages
have no answer. Both computers have the bus set at Wide (16-bit)
but each time Leicester, the second Precision, boots up each
drive is queried and changed to narrow. Odd, but no cure, at
this time.
Downloaded new firmware for the
LSI adapters (one on board, and one on the PCI bus). This was
purportedly successful: except for the "ERROR! Adapter
Malfunctioning" that comes up. The LSI PCI adapter seemingly is
gone away. After a few reboots, the LSI software completed, and
then the system OS, XP Pro SP2, came up. It immediately required
a found new hardware reboot.
Several firmware upgrade/cure
attempts unsuccessful. Now what? New U320 Adaptec adapter? It is
the PCI adapter, not the onboard port, that won't work.
Equally, it is the onboard connector that will only run at
narrow speeds (40MB).
Sunday, May 18th., 2008:
LSI firmware downloaded and installed. Now both of the LSI U320
adapters work, but only at narrow speeds. I will have to try a
different cable, and if that doesn't work, then I don't know
what will.
Using the LSI CIMBrowser, I have
discovered that the onboard chip has firmware revision
01.01.24.00, whereas the PCI adapter has version 01.03.52.00. I
shall have to use a different cable, but that may provide an
answer to the narrow rate problem.
Computers are strange, with
software installed on one machine producing different results on
another. The CIMBrowser, identical installations on the two Dell
Precisions, shows two (three connectors) adapters on the
troublesome machine, but only one on the properly working one.
Even a reboot and a reinstall gave the same result. Never mind,
I need to find a U320 terminating resistor and put it on a
cable, to see if I can improve the situation.
Tuesday, May 20th., 2008:
Here are a few images of two Dell computer device managers and
an associated set of the LSI U320 adapters installed in each. Note the LSI revisions. All three adapters, one of which is the on-board chip, differ. Now, it worsens: I removed the Leicester PCI LSI U320 adapter, because I wanted to determine if I could update the firmware on the computer's on-board chip. The machine would not produce video, although other activity was apparent. The cable and PCI card were reinstalled, and now the problem remains with no video apparent. Ridiculous situation, there are no beeps coming out of the machine, so it presumably believes itself healthy. Now what? Replacement is preferable, methinks. Wednesday, May 21st., 2008: Well, it was not any of the adapters: the problem is the aging of the on-board cable. Because it cools and warms and because it has a really hard fold in it, which those of you who have such a machine should be aware of, it has perished. I took the recalcitrant machine, Leicester, down to CSH, and we played around with it. Ultimately, we discovered that the cable was faulty by putting the shop's cable into the computer. Immediately, everything worked as it should. So, I await the arrival of the terminating resistor, and when I place it in the cable that I have, then everything should work properly. It means that I should also replace the cable on the "good" Dell fairly soon. This is the second time that this model made by Dell has acted as if it were a laptop. I must, repeat must, _always_ check the damned cables before doing anything at all to do with physical problems that arise on my computers. We had, at CHS, Computer Supply House, changed the motherboard, checked the connectors, and pratted about uselessly. When Dave suggested using the on-board cable on the adapter, which has two connectors, the problem continued, and was, it appeared, worse. When we placed the round workshop SCSI multi-connector cable in the Dell, then the LSI BIOS awoke and told us everything worked as it should. Wonder of wonders. Thursday, May 22nd., 2008: One should double check before closing the case: I checked the properly working Dell, Monmouth, today, to see what condition the on-board cable was in. It seemed fine, in a lot better condition than the faulty one in Leicester. When I replaced the connectors, I thought that they were properly fastened. This proved wrong, since one of them was not quite tight. This produced lots of errors with the computer. I had to shut down and go over every SCSI connection once more, and this did the trick. Only one LSI adapter is shown, nonetheless. Saturday, May 24th., 2008: New terminating resistor for the cable attached to the on-board LSI on Leicester works as required. Only one of the LSI chips is needed, since the cable takes more than the number of drives in the machine. That means I have a spare LSI 53C1030 PCI-X card. I tried to use it in the ML570, but it is not compatible with the drive cage installed therein, which is U160 based. The machine kept starting up into MS 2003 R2 Ent and then rebooting: not too good. So, back to the Adaptec 29160n, and that works well enough. Tuesday, May 27th., 2008: A moderator, Sorinso, on the Petri Forums, mentioned that he uses Minicom KVM. I have been looking at their products on their site, and they have what is called Smart IP Access, which includes CAT5 IP cable connection. That means, I think, that I could use their equipment on my two Domain Controllers that currently cannot be used with ordinary KVM switches. Something to think about. Wednesday, May 28th., 2008: I have been having perpetual synchronisation with off-line folder problems with Outlook. I have tried everything that I have found through various forums, including removing My Documents settings. Nothing has worked, so I have installed Thunderbird, which imported most of my settings and all of my email folders. I have been having problems setting up the securemail settings. If that is completed then I shall not be using Outlook any more. Somewhere there is a flaw: I can
access the MyHosting WebMail applet with the correct password. I can
open Outlook and it will receive. I cannot make Thunderbird accept
the password, even though opening Outlook and resetting the password
worked therein. With regard to KVM problems see this on the HP forums: ML570 G1 array config utility. It refers to the inability to have the ML570 BIOS seen by the KVM software, even those that use IP via CAT5 cable, such as items available from MiniCom. Thursday, May 29th., 2008: The problem was simply that, unlike in Outlook, one does not check the Use secure authentication in Thunderbird allied to the port change. When I unchecked that in the relevant accounts, for which see the third image above, everything worked. Friday, May 30th., 2008: I have installed/updated MS Office to 2007. Immediately after starting Outlook it asked me to install Instant Search. I decided not to, because I think that that may have been the culprit. We shall see. I had to hard reset the computer after returning from today's journeying. The clock had remained frozen at the time early this morning that Outlook 2003 had begun synchronising, and was still doing so and preventing any other action on the computer from executing. That at six in the evening. Hence I have been updating to Office 2007 on all, but one, of the relevant computers. I am just writing on another page, and up comes a statement from the task bar that Outlook has completed a data check? What? It means that it thinks somewhere there is a problem in a .PST file. Scanpst.exe to the rescue? Saturday, May 31st., 2008: Crossing my fingers slows my keyboard typing, but perhaps I need not. Outlook seems to be working properly again. There is a bar on the main page that irritates me: it wants me to install Instant Search. I don't think so: I would rather use the OS version, that I have returned to the Windows 2000 style. That was done to facilitate efficiency, because the latest style is all show, no knickers. Sunday, June 1st., 2008: Outlook has remained operational, although when typing a message it often hangs. One has to wait before it becomes usable once more. I shall have to obtain a round U320 cable for Monmouth, this double-headed box. I had trouble with the flat cable working itself loose under pressure and creating problems with one drive in particular. The cable is stressed because there is no straight run throughout, which means that its strength works against maintaining proper connections. The first two drives are 68-pin, and are followed by three SCA items. The connectors do not remain in line, and any knocks on the case, or other matters, will produce the results I had. When I had moved one of the drives to another slot, and given it the PCI-X LSI instead of the on-board chip, and put the other drive onto the on-board connector, everything started up as normal. The latest Excel, version 2007, that I am now using instead of 2003, is a real dog. The warnings that persist, even after one has chosen to override them, and the replacement of files that one is using when using Save As, are totally frustrating. In Excel 2003 one could save the .xls file as .htm without bother. Now, one has to click further on in the mess and it replaces the file that was being saved. Then, when one opens that file, data warnings appear once more. It is not natural to use, is Excel 2007, and that is ridiculous. It's late Sunday, and Outlook is locked again. What is it this time? Same or different? I have no clue. But it can't continue: I shall have to look for a different email programme, this is ridiculous. To have one's main workstation locked periodically in the past two or so weeks, when it never happened before implies that an update might have caused this, but how would I know which one? And, why would it affect both Outlook 2003 and 2007? Wednesday, June 4th., 2008: Of late, Outlook has been working well, except that when it starts to check for emails I worry whether it will complete or stall. Paranoid, what? I have had a few emails about Great Deals, run by that guy Jas. Failing drives, etc. The thing is, I thought that he had vanished, but his name is on lots of 73GB U320 drives for sale on eBay. I had thought that he wouldn't be back, but he probably thinks that he can continue to sell this ModusLnk/Fujitsu product with impunity. Friday, June 6th., 2008: Spoke too soon: Outlook solidly locked the email computer again: cannot access it through the network either. Hard reboot required. Will have to think about a decent replacement for it, but not Thunderbird. Monday, June 9th., 2008: The problem with Monmouth, the Dell Precision 650 that runs Outlook, may have been cured. It would seem that memory demands clash between Outlook and Firefox. The latter, of any version, has known memory leaks should one leave the programme running for noticeable lengths of time. Outlook has really heavy memory demands and it would seem that this could be the reason the computer eventually freezes. When that happens one cannot enter the three-fingered salute to determine how memory is being used: one cannot do anything. I have Firefox 3 RC2 running on other computers, and I hardly ever shut it down. It continues to work normally, but not on Monmouth alongside Outlook. QED. Thursday, June 12th., 2008: Well, it's not Firefox with Outlook that seizes/locks Monmouth, it is Outlook alone on its usual send/receive signal, where it decides its had enough work for the day and goes on strike. Bloody thing. This after yesterday's monthly MS update, which was mainly security for XP. I do not at this time want to update to Vista and see what if anything would change. If this is a flaky computer, which I doubt however, it won't be improved with merely an OS change. But, Linux? No, thanks. Tuesday, June 17th., 2008:
Yesterday, I opened a support ticket with Microsoft:
after having the techie in India control my computer for a while, we
determined that I could under Tools, Trust Center open the Add-ins.
This for the first time for me: it had stuck on this push button
every other time I had tried: I thought that that indicated there
were no such things active. Wednesday, June 18th., 2008: Unfortunately, having left Outlook and Firefox running overnight, the computer again seized when I tried to use the former this morning. Hard reboot once more. This does the computer no good. Thursday, June 19th., 2008: Yes, still worried about Outlook 2007. However, it is still working after another day, which I believe is related to the fact that I did not have Firefox active. IE 7 was active, but that does not leak like Firefox does, it would seem. Will have Outlook constantly active and wait. So, after being out and clicking on the Outlook icon on the running application bar, it opened and seemed to be working. Except that clicking on an email promptly made it hang. Eventually, after a couple of three-fingered salutes the machine came back to life. This is the first time that a stall has not required a hard reboot. Maybe I should copy the pst file over to another machine and make that the Outlook box to see what happens with it. Will decide about that in due course. Saturday, June 21st., 2008: Another hard reboot day. Outlook opened when clicked, started to open the emails that had arrived overnight and then slowed down. Nothing on the task or application bar would open, neither would anything on the desktop. Mouse could be moved around but couldn't do anything. Three-fingered salute failed to complete. Button pressing time. Then, when the computer, Monmouth (one of the two Dell Precision 650 computers here) was up Outlook started fine, but did not seemingly require a pst check. This is strange, since every other hard reboot has resulted in Outlook checking the file for errors. I have emailed the MS support team, and related to them that a Petri Outlook reference page has proved useful, but the problems with this important programme have not disappeared. I have also opened a forum posting on the XP pages on the Petri site. It's beneficial to have others helping, because I don't want to give up on this significant snafu. Monday, June 23rd., 2008:
This is strange, the programme is working, and has done for almost
two days now. Also, other programmes, including Firefox, seem to be
more responsive. Tuesday,
June 24th., 2008: Additional information that means something is
wrong in the Dell arena. Thursday, July 3rd., 2008: First day in some time that the recalcitrant computer still works in the morning. I removed the U320 LSI PCI adapter yesterday, and also one hard drive, since the cable won't take five drives. Having checked the physical connections I restarted the machine and it has worked without fault for twenty-four hours. Really odd, and perhaps having two LSI U320 chips loaded caused problems. I hope it was that, and that the computer continues to run properly. Time will tell, indeed. I shall, eventually, return the Outlook PST file to the Dell, just to see if it will perform without seizing again. If that's the case, then I found the cause of all the recent freezing that caused so many hard reboots. Friday, July 4th., 2008:
Well, I don't think it was the LSI, because the same halting,
stumbling performance kept on coming. But, having installed Acronis
software, there were, in consequence, two services installed that
came along. These I have disabled. Since then, nothing untoward has
happened. It may be only a few hours, but I really think that this
is a significant step. Firefox 3 RC 4 crashed on Leicester, the
other Dell Precision, which is a change, and on the Telegraph
newspaper site, which isn't. Peculiar, what? |
