probook 6560b does not recognize my SSD

[Dusano]

I didn't make any performence tests yet, but here are some reports from linux.

The SSD is visible in linux - SmartCTL report:

=== START OF INFORMATION SECTION ===
Device Model:     OCZ-AGILITY3
Firmware Version: 2.15
User Capacity:    120,034,123,776 bytes [120 GB]
Sector Size:      512 bytes logical/physical
Device is:        Not in smartctl database [for details use: -P showall]
ATA Version is:   8
ATA Standard is:  ACS-2 revision 3
Local Time is:    Tue Dec 20 21:19:03 2011 UTC
SMART support is: Available - device has SMART capability.
SMART support is: Enabled 

When I checked some logs I found this in dmesg:

[    8.917555] ata1: SATA link up 6.0 Gbps (SStatus 133 SControl 300)
[    8.928839] ata1.00: failed to IDENTIFY (I/O error, err_mask=0x100)
[    8.928849] ata1: limiting SATA link speed to 3.0 Gbps
[   14.221839] ata1: SATA link up 3.0 Gbps (SStatus 123 SControl 320)
[   14.233394] ata1.00: ACPI cmd 00/00:00:00:00:00:a0 (NOP) rejected by device (Stat=0x51 Err=0x04)
[   14.234247] ata1.00: ATA-8: OCZ-AGILITY3, 2.15, max UDMA/133
[   14.234254] ata1.00: 234441648 sectors, multi 16: LBA48 NCQ (depth 31/32), AA
[   14.243348] ata1.00: ACPI cmd 00/00:00:00:00:00:a0 (NOP) rejected by device (Stat=0x51 Err=0x04)
[   14.244205] ata1.00: configured for UDMA/133
[   14.244391] scsi 0:0:0:0: Direct-Access     ATA      OCZ-AGILITY3     2.15 PQ: 0 ANSI: 5 

It seems that linux is decreasing communication speed from 6 Gbps to 3 Gbps due to some problem and finally is configured for UDMA/133. I do not understand everything in this log, but maybe you guys do.

I also upgraded BIOS to version F20, but nothing changed.

[Slym]

This is exactly the kind of errors I was talking about in my previous post (Limiting speed of SATA Link + errors).

It's very bad because this error is causing boot to be (a lot) longer than it should.

PS : Every single SSD I tried on eSATA link worked fine. (SATA3 and SATA2)

[6560b_owner]

Thanks for the extended log.

Linux tries to contact the drive via sata-III (6gb/s) and fails.

One 5s timeout later, linux retries via sata-II (3gb/s) and is sucessfull.

(IMHO, this timeout can be avoided by certain options in linux)

Due to the 8b/10b code (checksum etc.), you have a theoretical troughput of 300MB/s on sata-II. 

Not bad as a fallback mode. UDMA/133 though does not transmit more than 133MB/s, so I'm not sure which value is correct.

A quick benchmark with hdparm (shipped with most linux distributions) would answer that question.

You said, that although the SSD drive is recognized via esata, booting from it is impossible. Did you activate the relevant bios options (esata boot/boot priority)?

[Slym]

I did not say that booting on eSATA won't work. It's actually working pretty well.

I used LinuxLive USB Creator (my own software) to create dozens of Live "USB" on my eSATA SSD and it's booting really fast ! 

I even tried Windows 8 "To Go" (kind of a live USB) and it worked fine too.

[6560b_owner]

@Slym:

That's good news then.

If the firmware/EFI/BIOS-thing of our hp notebooks is able to boot from an external sata-III ssd via esata, that means there is almost nothing wrong with it. Esata is practically sata with a different cable connector.

Some more log details from linux that contradict my previous informations about the sata connection of the external drive port.

Perhaps that helps someone  to figure the whole thing out.

---

Kernel: 3.0.0.12, 64 Bit
Output of 'lshw' run as root
My 6560b is about 3 months old. HW65 chipset / board

    description: Notebook
    product: HP ProBook 6560b (LG658EA#ABD)
[..]

     *-firmware
          description: BIOS
          vendor: Hewlett-Packard
          physical id: b
          version: 68SCE Ver. F.20
          date: 09/29/2011
[..]   
    *-storage
             description: SATA controller
             product: 6 Series/C200 Series Chipset Family 6 port SATA AHCI Controller
             vendor: Intel Corporation
             physical id: 1f.2
             bus info: pci@0000:00:1f.2
             logical name: scsi0 // comment: main drive
             logical name: scsi1 // comment: optical drive bay
             logical name: scsi4 // external esata drive
             version: 04
 [..]

then, all 3 sata-drives (main, optical and external) are listed, including their partitions.

---

-the main drive slot is enumerated as 'scsi 0', that means port 0. (Linux disk utility: 'Port 1')
-the optical drive slot (=upgrade bay) is enumerated as 'scsi 1', that means port 1. (Linux disk utility: 'Port 2')

-the external drive (=esata) is enumerated as 'scsi 4', that means port 4. (Linux disk utility: 'Port 5')

To me, that clearly indicates that the Linux disk utility just wants to spare the regular user from seeing  the zero-based counting  that is usual in computing.

According to this data, both internal drive slots are on the first ports that are linked via sata-III connections (according to informations about the chipset elsewhere).

This interpretation could explain two things:

both internal drive connections don't work with sata-III while the external one does

because the external connection is sata-II?

as well as

the external drive does only work in sata-II speed

.. same answer: because the external connection is sata-II?

What do you think of the external sata connections/the implications?

By the way: did any of you try to contact support yet? 

We have a reproducable, isolated, severe malfunction.

We know which drives won't work. We know on which ports they don't work.

We have detailled information about our used firmware and hardware.

We are first/second google hits on '6560b malfunction' and  '6560b faulty' ('6560b defective':  third) - which is bad enough for HP if they continue to ignore the bug (or, more optimistically: haven't been noticed yet)

Experienced users will search for problems with the notebook before ordering one and refrain from buying it.

I just saw that this thread is placed under the category Notebook > Upgrades. That probably means that HP staff will seldomly check this part of the forum.

I'll try to open a new thread in Notebook>Locks, Freezes & Hangs and link here.

[Update] Well, here it is.

[Update2] I just tried to contact the frequently posting HP employee 'kamat_hp' via PM, but that was blocked. Any other hp employee I have seen here so far is fairly inanctive (not online since weeks).

I'll try to contact one of the few remaining HP employees here.

[Update3] The pm to the hp employee SiobhanF was sent sucessfully.

[Slym]

I opened a support case from the Active Chat Support but it was useless. They did not understand my request.

They asked me to flash a new firmware revision on my disk but the firmware was for the official Intel SSD (mine was Crucial M4) so this was just a stupid suggestion 😕

I lost my time. Maybe someone else should try to open a support case.

[6560b_owner]

@Slym: thanks for your efforts. Perhaps christmas was not the best time to contact live support ..

Concerning the sata-port numbers:

I've confirmed my findings under linux independently with the Intel rapid storage tool under win 7 (x64bit).

The port numbers match - port 0 for the main hdd, port 1 for the upgrade bay / optical drive and port 4 for the esata port.

Interestingly, port 3 and 5 are also marked as 'external' - very likely, these ports aren't used though.

why this is important:

If the port enumeration matches the pattern of the older steppings (that were affected with the intel-sata-II-bug), then port 0 and 1 are sata-III while the others are not. (mentioned here, for example)

Which means (I'm repeating myself, but I wasn't sure before) that sata-III drives are only a problem when connected to the sata-III ports.

That is probably the reason why the upgrade bay (port 1, also sata-III) does not like sata-III drives as well.

[Slym]

My instant support session was open on 08/08/2011 not on christmas lol.

[6560b_owner]

@Slym: that does not look good for the hp live support..

Could you please try again via phone or email? We have to get some attention from somebody at HP, one way or another..

Some news. I finally received my second Kingston HyperX 120GB (sata-III, speeds up to 500MB/s) as a replacement for my broken OCZ Agility 2.

I connected the HyperX directly via Esata (special connector required as esata and sata are electrically identical, but have different cable connectors) and sucessfully booted from it (6560b).

That possibility is definitely confirmed then.

Benchmarks done under linux (ext4 partition on the ssd) with the ssd connected to the 6560b via esata show up to 270Mb read performance, indicating that the e-sata connection is sata-II, which is what we anticipated.

The drive does signal general sata-III (6Gb/s) capability, but is run in ultra-dma mode 6 according to hdparm.

Due to an known chipset issue with ssd + ahci bios on some amd 750 chipsets, which prevents ssds to be bootable with AHCI enabled, I looked into that as a possible cause of our troubles, too. But the 6560b notebook run in IDE mode (instead of AHCI) does still not recognize the HyperX drive. Therefore, IDE/AHCI is clearly not the issue here.

[jlianes]

This problem really annoys me. I have bough in 6 6560b laptops. First I tried with Force 3 SSD drives. HAs never worked. Then I read on Crucial's page that the M4 was garanteed to work so I have used 3 240GB M4 drives and all of them have worked fine (firmware revision 9) Two of the computers I needed to take apart the hdd caddy to get the ssd to get a little closer into the SATA port before it worked... strange....

Well now I have a 120GB M4 for the first time, and it's not working. Firmware revision 9 and Bios revision F20....