 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Yesterday a drive in the RAID5 array that this server is stored on failed*.
This reduced the three-disk array to a non-redundant two-disk array with a
hot spare. While the RAID controller was initializing the hot spare disk, a
second drive failed, thus destroying the array and taking povray.org offline.
I was able to boot the system off a backup disk which had a copy of the
system data ranging from 24 hours to about 8 hours old; the newsgroup data
was probably about 12-18 hours old. This is the disk the system is running
off at the time of this post.
Thus, postings to the newsgroups made since the backup was taken and up until
the time of the crash will be missing. If you posted recently, you may want
to ensure the article is still on the server (just because you see it in your
news reader does not mean that it's physically on the server - the means of
determining this varies from application to application).
-- Chris
* the controller (Adaptec 1210SA) claimed the drives had failed. however
subsequent examination casts doubt on this. this controller has been
nothing but trouble since it was purchased and has been the primary source
of all machine downtime for the past few years. I will not be purchasing
Adaptec again.
NB fans etc have been checked; the machine is running cool and the PSU is
fine. we can't find a legitimate fault with the disks or chassis.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I am re-posting this message several times in order to bump the message
number for this group to at least what it would have been before the data loss.
Note: some news reader programs cache the highest item number seen for each
newsgroup and upon the next visit only ask for messages higher than that
number (this only affects those using NNTP to access the group, of course,
rather than the web interface). Hence, unless we bump the internal message
number to at least the old value, people using those readers won't see some
of the first messages posted after the restore, including this one. Hence my
re-posting.
Chris Cason wrote:
> Yesterday a drive in the RAID5 array that this server is stored on failed*.
> This reduced the three-disk array to a non-redundant two-disk array with a
> hot spare. While the RAID controller was initializing the hot spare disk, a
> second drive failed, thus destroying the array and taking povray.org offline.
>
> I was able to boot the system off a backup disk which had a copy of the
> system data ranging from 24 hours to about 8 hours old; the newsgroup data
> was probably about 12-18 hours old. This is the disk the system is running
> off at the time of this post.
>
> Thus, postings to the newsgroups made since the backup was taken and up until
> the time of the crash will be missing. If you posted recently, you may want
> to ensure the article is still on the server (just because you see it in your
> news reader does not mean that it's physically on the server - the means of
> determining this varies from application to application).
>
> -- Chris
>
> * the controller (Adaptec 1210SA) claimed the drives had failed. however
> subsequent examination casts doubt on this. this controller has been
> nothing but trouble since it was purchased and has been the primary source
> of all machine downtime for the past few years. I will not be purchasing
> Adaptec again.
>
> NB fans etc have been checked; the machine is running cool and the PSU is
> fine. we can't find a legitimate fault with the disks or chassis.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I am re-posting this message several times in order to bump the message
number for this group to at least what it would have been before the data loss.
Note: some news reader programs cache the highest item number seen for each
newsgroup and upon the next visit only ask for messages higher than that
number (this only affects those using NNTP to access the group, of course,
rather than the web interface). Hence, unless we bump the internal message
number to at least the old value, people using those readers won't see some
of the first messages posted after the restore, including this one. Hence my
re-posting.
Chris Cason wrote:
> Yesterday a drive in the RAID5 array that this server is stored on failed*.
> This reduced the three-disk array to a non-redundant two-disk array with a
> hot spare. While the RAID controller was initializing the hot spare disk, a
> second drive failed, thus destroying the array and taking povray.org offline.
>
> I was able to boot the system off a backup disk which had a copy of the
> system data ranging from 24 hours to about 8 hours old; the newsgroup data
> was probably about 12-18 hours old. This is the disk the system is running
> off at the time of this post.
>
> Thus, postings to the newsgroups made since the backup was taken and up until
> the time of the crash will be missing. If you posted recently, you may want
> to ensure the article is still on the server (just because you see it in your
> news reader does not mean that it's physically on the server - the means of
> determining this varies from application to application).
>
> -- Chris
>
> * the controller (Adaptec 1210SA) claimed the drives had failed. however
> subsequent examination casts doubt on this. this controller has been
> nothing but trouble since it was purchased and has been the primary source
> of all machine downtime for the past few years. I will not be purchasing
> Adaptec again.
>
> NB fans etc have been checked; the machine is running cool and the PSU is
> fine. we can't find a legitimate fault with the disks or chassis.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I am re-posting this message several times in order to bump the message
number for this group to at least what it would have been before the data loss.
Note: some news reader programs cache the highest item number seen for each
newsgroup and upon the next visit only ask for messages higher than that
number (this only affects those using NNTP to access the group, of course,
rather than the web interface). Hence, unless we bump the internal message
number to at least the old value, people using those readers won't see some
of the first messages posted after the restore, including this one. Hence my
re-posting.
Chris Cason wrote:
> Yesterday a drive in the RAID5 array that this server is stored on failed*.
> This reduced the three-disk array to a non-redundant two-disk array with a
> hot spare. While the RAID controller was initializing the hot spare disk, a
> second drive failed, thus destroying the array and taking povray.org offline.
>
> I was able to boot the system off a backup disk which had a copy of the
> system data ranging from 24 hours to about 8 hours old; the newsgroup data
> was probably about 12-18 hours old. This is the disk the system is running
> off at the time of this post.
>
> Thus, postings to the newsgroups made since the backup was taken and up until
> the time of the crash will be missing. If you posted recently, you may want
> to ensure the article is still on the server (just because you see it in your
> news reader does not mean that it's physically on the server - the means of
> determining this varies from application to application).
>
> -- Chris
>
> * the controller (Adaptec 1210SA) claimed the drives had failed. however
> subsequent examination casts doubt on this. this controller has been
> nothing but trouble since it was purchased and has been the primary source
> of all machine downtime for the past few years. I will not be purchasing
> Adaptec again.
>
> NB fans etc have been checked; the machine is running cool and the PSU is
> fine. we can't find a legitimate fault with the disks or chassis.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I am re-posting this message several times in order to bump the message
number for this group to at least what it would have been before the data loss.
Note: some news reader programs cache the highest item number seen for each
newsgroup and upon the next visit only ask for messages higher than that
number (this only affects those using NNTP to access the group, of course,
rather than the web interface). Hence, unless we bump the internal message
number to at least the old value, people using those readers won't see some
of the first messages posted after the restore, including this one. Hence my
re-posting.
Chris Cason wrote:
> Yesterday a drive in the RAID5 array that this server is stored on failed*.
> This reduced the three-disk array to a non-redundant two-disk array with a
> hot spare. While the RAID controller was initializing the hot spare disk, a
> second drive failed, thus destroying the array and taking povray.org offline.
>
> I was able to boot the system off a backup disk which had a copy of the
> system data ranging from 24 hours to about 8 hours old; the newsgroup data
> was probably about 12-18 hours old. This is the disk the system is running
> off at the time of this post.
>
> Thus, postings to the newsgroups made since the backup was taken and up until
> the time of the crash will be missing. If you posted recently, you may want
> to ensure the article is still on the server (just because you see it in your
> news reader does not mean that it's physically on the server - the means of
> determining this varies from application to application).
>
> -- Chris
>
> * the controller (Adaptec 1210SA) claimed the drives had failed. however
> subsequent examination casts doubt on this. this controller has been
> nothing but trouble since it was purchased and has been the primary source
> of all machine downtime for the past few years. I will not be purchasing
> Adaptec again.
>
> NB fans etc have been checked; the machine is running cool and the PSU is
> fine. we can't find a legitimate fault with the disks or chassis.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I am re-posting this message several times in order to bump the message
number for this group to at least what it would have been before the data loss.
Note: some news reader programs cache the highest item number seen for each
newsgroup and upon the next visit only ask for messages higher than that
number (this only affects those using NNTP to access the group, of course,
rather than the web interface). Hence, unless we bump the internal message
number to at least the old value, people using those readers won't see some
of the first messages posted after the restore, including this one. Hence my
re-posting.
Chris Cason wrote:
> Yesterday a drive in the RAID5 array that this server is stored on failed*.
> This reduced the three-disk array to a non-redundant two-disk array with a
> hot spare. While the RAID controller was initializing the hot spare disk, a
> second drive failed, thus destroying the array and taking povray.org offline.
>
> I was able to boot the system off a backup disk which had a copy of the
> system data ranging from 24 hours to about 8 hours old; the newsgroup data
> was probably about 12-18 hours old. This is the disk the system is running
> off at the time of this post.
>
> Thus, postings to the newsgroups made since the backup was taken and up until
> the time of the crash will be missing. If you posted recently, you may want
> to ensure the article is still on the server (just because you see it in your
> news reader does not mean that it's physically on the server - the means of
> determining this varies from application to application).
>
> -- Chris
>
> * the controller (Adaptec 1210SA) claimed the drives had failed. however
> subsequent examination casts doubt on this. this controller has been
> nothing but trouble since it was purchased and has been the primary source
> of all machine downtime for the past few years. I will not be purchasing
> Adaptec again.
>
> NB fans etc have been checked; the machine is running cool and the PSU is
> fine. we can't find a legitimate fault with the disks or chassis.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I am re-posting this message several times in order to bump the message
number for this group to at least what it would have been before the data loss.
Note: some news reader programs cache the highest item number seen for each
newsgroup and upon the next visit only ask for messages higher than that
number (this only affects those using NNTP to access the group, of course,
rather than the web interface). Hence, unless we bump the internal message
number to at least the old value, people using those readers won't see some
of the first messages posted after the restore, including this one. Hence my
re-posting.
Chris Cason wrote:
> Yesterday a drive in the RAID5 array that this server is stored on failed*.
> This reduced the three-disk array to a non-redundant two-disk array with a
> hot spare. While the RAID controller was initializing the hot spare disk, a
> second drive failed, thus destroying the array and taking povray.org offline.
>
> I was able to boot the system off a backup disk which had a copy of the
> system data ranging from 24 hours to about 8 hours old; the newsgroup data
> was probably about 12-18 hours old. This is the disk the system is running
> off at the time of this post.
>
> Thus, postings to the newsgroups made since the backup was taken and up until
> the time of the crash will be missing. If you posted recently, you may want
> to ensure the article is still on the server (just because you see it in your
> news reader does not mean that it's physically on the server - the means of
> determining this varies from application to application).
>
> -- Chris
>
> * the controller (Adaptec 1210SA) claimed the drives had failed. however
> subsequent examination casts doubt on this. this controller has been
> nothing but trouble since it was purchased and has been the primary source
> of all machine downtime for the past few years. I will not be purchasing
> Adaptec again.
>
> NB fans etc have been checked; the machine is running cool and the PSU is
> fine. we can't find a legitimate fault with the disks or chassis.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I am re-posting this message several times in order to bump the message
number for this group to at least what it would have been before the data loss.
Note: some news reader programs cache the highest item number seen for each
newsgroup and upon the next visit only ask for messages higher than that
number (this only affects those using NNTP to access the group, of course,
rather than the web interface). Hence, unless we bump the internal message
number to at least the old value, people using those readers won't see some
of the first messages posted after the restore, including this one. Hence my
re-posting.
Chris Cason wrote:
> Yesterday a drive in the RAID5 array that this server is stored on failed*.
> This reduced the three-disk array to a non-redundant two-disk array with a
> hot spare. While the RAID controller was initializing the hot spare disk, a
> second drive failed, thus destroying the array and taking povray.org offline.
>
> I was able to boot the system off a backup disk which had a copy of the
> system data ranging from 24 hours to about 8 hours old; the newsgroup data
> was probably about 12-18 hours old. This is the disk the system is running
> off at the time of this post.
>
> Thus, postings to the newsgroups made since the backup was taken and up until
> the time of the crash will be missing. If you posted recently, you may want
> to ensure the article is still on the server (just because you see it in your
> news reader does not mean that it's physically on the server - the means of
> determining this varies from application to application).
>
> -- Chris
>
> * the controller (Adaptec 1210SA) claimed the drives had failed. however
> subsequent examination casts doubt on this. this controller has been
> nothing but trouble since it was purchased and has been the primary source
> of all machine downtime for the past few years. I will not be purchasing
> Adaptec again.
>
> NB fans etc have been checked; the machine is running cool and the PSU is
> fine. we can't find a legitimate fault with the disks or chassis.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
|
|
