HP MicroServer ! Consiglio mini serverino / nas

Pagina 19 di 23
prima
... 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 ultimo
Visualizzazione dei risultati da 181 a 190 su 222
  1. #181

    Predefinito

    Originariamente inviato da squonk
    Scusami ma non ho capito.
    Supponiamo che io installi OMV su un disco da 2.5" (disco dedicato esclusivamente al SO, come indicato nella documentazione: non è possibile usare una partizione di un hard disc!) e metta tutti i dati sui dischi aggiuntivi (uno o più, configurati con o senza raid).
    Se si rompe il disco su cui è installato OMV, i dati diventano illeggibili? Per quale motivo? I file rimangono accessibili solo se si mantiene integro il sistema? Mi sembra davvero strano...
    No assolutamente, ma finchè non trovi un altro hdd e reinstalli OMV sei senza NAS e puoi leggere i dati solo collegando i dischi ad un altro pc. Con il DSM invece il sistema è replicato su tutti i dischi quindi qualsiasi hdd si rompa, il sistema non va' perso

  2. #182
    ●⁞◌ Ȏrȉzzȏntέ Ðέglȋ ȨvέntȊ ◌⁞●
    Registrato
    Aug 2008
    Località
    Palermo
    Messaggi
    2,952

    Predefinito

    La vulnerabilità che ho indicato era in profondità al sistema operativo usato da Synology e non era evitabile tramite
    alcun tipo di firewall

    Direttamente da SecurityFocus:

    "**************************************************************
    Title: Synology DSM multiple vulnerabilities
    Version affected: <= 4.3-3776
    Vendor: Synology
    Discovered by: Andrea Fabrizi
    Email: andrea.fabrizi (at) gmail (dot) com [email concealed]
    Web: Andrea Fabrizi
    Twitter: @andreaf83
    Status: unpatched
    **************************************************************

    Synology DiskStation Manager (DSM) it's a Linux based operating
    system, used for the DiskStation and RackStation products.

    1] ======== Remote file download ========
    Any authenticated user, even with the lowest privilege, can download
    any system file, included the /etc/shadow, samba password files and
    files owned by the other DSM users, without any restriction.

    The vulnerability is located in "/webman/wallpaper.cgi". The CGI takes
    as parameter the full path of the image to download, encoded in ASCII
    Hex format.
    The problem is that any file type can be downloaded (not only images)
    and the path validation is very poor. In fact the CGI checks only if
    the path starts with an allowed directory (like
    /usr/syno/synoman/webman), and this kind of protection can be easily
    bypassed using the ../ attack.

    For example to access the /etc/shadow:
    2f7573722f73796e6f2f73796e6f6d616e2f7765626d616e2f2e2e2f2e2e2f2e2e2f2e2e
    2f6574632f736861646f77
    (/usr/syno/synoman/webman/../../../../etc/shadow)

    ------------------------------------------
    GET /webman/wallpaper.cgi?path=AABBCCDDEEFF11223344 HTTP/1.1
    Host: 127.0.0.1:5000
    Cookie: stay_login=0; id=XXXXXXXXXXX
    ------------------------------------------

    2] ======== Command injection ========
    A command injection vulnerability, present on the
    "/webman/modules/ControlPanel/ modules/externaldevices.cgi" CGI,
    allows any administrative user to execute arbitrary commands on the
    system, with root privileges.

    ------------------------------------------
    POST /webman/modules/ControlPanel/modules/externaldevices.cgi HTTP/1.1
    Host: 127.0.0.1:5000
    User-Agent: ls
    Cookie: stay_login=0; id=XXXXXXXXXXX
    Content-Length: 128

    action=apply&device_name=aa&printerid=1.1.1.1-aa';$HTTP_USER_AGENT>/tmp/
    output+%23&printer_mode=netPrinter&eject_netprinter=true
    ------------------------------------------

    Putting the command to execute as the User Agent string, after the
    request the output will be ready into the /tmp/output file.

    3] ======== Partial remote content download ========
    For the localization DSM uses some CGI, that takes the lang parameter
    (e.g. "enu" for english) and returns a Json object containing the
    localized strings in a dictionary format.

    The strings are taken from a local file with the following path:
    [current_dir]/texts/[lang_parameter_value]/strings

    The "/strings" appended at the end of the path prevents a path
    injection, because any value injected using the "lang" parameter will
    be invalidated (in other words, it's possible to read only files named
    "strings"). But, the interesting thing is that the full path of the
    strings files is built using a snprintf function like that:

    snprintf(&s, 0x80u, "texts/%s/strings", lang)

    This means that putting a lang value big enough, it's possible to
    overflow the 128 byte allowed by the snprintf and take out the
    "/strings" from the built path.

    For example, the lang value
    "./////////////////////////////////////////////////////////////////////
    ///////////////////../../../../../etc/synoinfo.conf" allow to get the
    /etc/synoinfo.conf file content.

    The second problem is that the input file taken by the CGI must be
    formatted in a key/value way: key1=string1

    In other words, to get some content from a generic file it's necessary
    that the file contains at least an "=" for each line (this is the
    reason why I called the vulnerability "Partial remote content
    download").

    At first glance it may seems very limiting, but, seen that it's
    possible to read directly from the disk block device (e.g.
    /dev/vg1000/lv), the amount of data dumped is very huge. In my tests I
    was able to dump around the 25/30% of the drive (tested with mixed
    content, like documents, images, generic files). It's possible to dump
    data from any drive connected. Interesting data can be also dumped
    from the /proc vfs.

    This vulnerability impacts two different CGI and is exploitable
    without authentication by any remote user:

    /scripts/uistrings.cgi
    /webfm/webUI/uistrings.cgi

    ------------------------------------------
    GET /scripts/uistrings.cgi?lang=XXXXXXXXX HTTP/1.1
    Host: 127.0.0.1:5000
    ------------------------------------------

    In the system there are two other uistrings.cgi, but are not affected.

    4] XSS
    A classic Cross-site scripting affects the following CGI:
    /webman/info.cgi?host=XXXX&target=XXXX&add=XXXX
    "


    Certo, facendo riferimento ai pericoli del passato, a preoccuparsi devono essere solo una parte degli utenti Synology.

    Detto questo però, considerato il livello e la qualità dello sviluppo del software della batteria di ingegneri di Synology,
    facevo invece riferimento più che altro alle preoccupazioni di quanto possa (verosimilmente) continuare ad accadere
    anche in futuro

  3. #183

    Predefinito

    Originariamente inviato da Totocellux
    La vulnerabilità che ho indicato era in profondità al sistema operativo usato da Synology e non era evitabile tramite
    alcun tipo di firewall:

    "**************************************************************
    Title: Synology DSM multiple vulnerabilities
    Version affected: <= 4.3-3776
    Vendor: Synology
    Discovered by: Andrea Fabrizi
    Email: andrea.fabrizi (at) gmail (dot) com [email concealed]
    Web: Andrea Fabrizi
    Twitter: @andreaf83
    Status: unpatched
    **************************************************************

    Synology DiskStation Manager (DSM) it's a Linux based operating
    system, used for the DiskStation and RackStation products.

    1] ======== Remote file download ========
    Any authenticated user, even with the lowest privilege, can download
    any system file, included the /etc/shadow, samba password files and
    files owned by the other DSM users, without any restriction.

    The vulnerability is located in "/webman/wallpaper.cgi". The CGI takes
    as parameter the full path of the image to download, encoded in ASCII
    Hex format.
    The problem is that any file type can be downloaded (not only images)
    and the path validation is very poor. In fact the CGI checks only if
    the path starts with an allowed directory (like
    /usr/syno/synoman/webman), and this kind of protection can be easily
    bypassed using the ../ attack.

    For example to access the /etc/shadow:
    2f7573722f73796e6f2f73796e6f6d616e2f7765626d616e2f2e2e2f2e2e2f2e2e2f2e2e
    2f6574632f736861646f77
    (/usr/syno/synoman/webman/../../../../etc/shadow)

    ------------------------------------------
    GET /webman/wallpaper.cgi?path=AABBCCDDEEFF11223344 HTTP/1.1
    Host: 127.0.0.1:5000
    Cookie: stay_login=0; id=XXXXXXXXXXX
    ------------------------------------------

    2] ======== Command injection ========
    A command injection vulnerability, present on the
    "/webman/modules/ControlPanel/ modules/externaldevices.cgi" CGI,
    allows any administrative user to execute arbitrary commands on the
    system, with root privileges.

    ------------------------------------------
    POST /webman/modules/ControlPanel/modules/externaldevices.cgi HTTP/1.1
    Host: 127.0.0.1:5000
    User-Agent: ls
    Cookie: stay_login=0; id=XXXXXXXXXXX
    Content-Length: 128

    action=apply&device_name=aa&printerid=1.1.1.1-aa';$HTTP_USER_AGENT>/tmp/
    output+%23&printer_mode=netPrinter&eject_netprinter=true
    ------------------------------------------

    Putting the command to execute as the User Agent string, after the
    request the output will be ready into the /tmp/output file.

    3] ======== Partial remote content download ========
    For the localization DSM uses some CGI, that takes the lang parameter
    (e.g. "enu" for english) and returns a Json object containing the
    localized strings in a dictionary format.

    The strings are taken from a local file with the following path:
    [current_dir]/texts/[lang_parameter_value]/strings

    The "/strings" appended at the end of the path prevents a path
    injection, because any value injected using the "lang" parameter will
    be invalidated (in other words, it's possible to read only files named
    "strings"). But, the interesting thing is that the full path of the
    strings files is built using a snprintf function like that:

    snprintf(&s, 0x80u, "texts/%s/strings", lang)

    This means that putting a lang value big enough, it's possible to
    overflow the 128 byte allowed by the snprintf and take out the
    "/strings" from the built path.

    For example, the lang value
    "./////////////////////////////////////////////////////////////////////
    ///////////////////../../../../../etc/synoinfo.conf" allow to get the
    /etc/synoinfo.conf file content.

    The second problem is that the input file taken by the CGI must be
    formatted in a key/value way: key1=string1

    In other words, to get some content from a generic file it's necessary
    that the file contains at least an "=" for each line (this is the
    reason why I called the vulnerability "Partial remote content
    download").

    At first glance it may seems very limiting, but, seen that it's
    possible to read directly from the disk block device (e.g.
    /dev/vg1000/lv), the amount of data dumped is very huge. In my tests I
    was able to dump around the 25/30% of the drive (tested with mixed
    content, like documents, images, generic files). It's possible to dump
    data from any drive connected. Interesting data can be also dumped
    from the /proc vfs.

    This vulnerability impacts two different CGI and is exploitable
    without authentication by any remote user:

    /scripts/uistrings.cgi
    /webfm/webUI/uistrings.cgi

    ------------------------------------------
    GET /scripts/uistrings.cgi?lang=XXXXXXXXX HTTP/1.1
    Host: 127.0.0.1:5000
    ------------------------------------------

    In the system there are two other uistrings.cgi, but are not affected.

    4] XSS
    A classic Cross-site scripting affects the following CGI:
    /webman/info.cgi?host=XXXX&target=XXXX&add=XXXX
    "


    Certo, facendo riferimento ai pericoli del passato, a preoccuparsi devono essere solo una parte degli utenti Synology.

    Detto questo però, considerato il livello e la qualità dello sviluppo del software della batteria di ingegneri di Synology,
    facevo invece riferimento più che altro alle preoccupazioni di quanto possa (verosimilmente) continuare ad accadere
    anche in futuro
    Beh quello si, con qualsiasi sistema open anche se proprietario di un determinato produttore, quando finisce nelle mani dei comuni utenti cose del genere sono sempre dietro l'angolo. Il consiglio di aggiornare sempre all'ultima versione vale anche in questo caso

  4. #184
    ●⁞◌ Ȏrȉzzȏntέ Ðέglȋ ȨvέntȊ ◌⁞●
    Registrato
    Aug 2008
    Località
    Palermo
    Messaggi
    2,952

    Predefinito

    Originariamente inviato da x9drive9in
    [...]
    Il consiglio di aggiornare sempre all'ultima versione vale anche in questo caso
    si vale anche in questo caso, tant'è che sempre nello stesso thread indicato in precedenza del giugno di quest'anno,
    era già stato caldeggiato tramite la procedura adottata ufficialmente da Synology sul proprio sito web:

    http://www.synology.com/en-global/co...ws/article/437


  5. #185
    kibibyte
    Registrato
    May 2012
    Messaggi
    311

    Predefinito

    In merito alla discussione su Synology volevo cercare di capire meglio il senso dei vostri interventi.
    Totocellux ha indicato che i programmatori di Synology si sono mossi tardi rispetto a delle vulnerabilità note da tempo. Hanno poi tappato le falle con gli aggiornamenti.
    La necessità di tenere un sistema aggiornato all'ultima versione è (per me) sempre fuori discussione.
    Ma permettetemi di chiedervi di chiarire meglio il senso di questi discorsi in questa sede.
    Stiamo dicendo che il software di Synology è più sicuro di altri perchè ha una casa produttrice che ci sta dietro? Quindi sarebbe da preferire un progetto come XPEnology rispetto ad altri progetti open come FreeNAS e OMV?
    Vuol dire che progetti come gli ultimi citati non sono così sensibili alle questioni di sicurezza e vengono aggiornati con meno frequenza?
    Io ho sempre pensato il contrario...
    Sulla bontà del software Synology ho visto poche discussioni: in molti, quando si sceglie un NAS commerciale, optano per quest'ultimo per la bontà del DSM, ma non è detto che gli sviluppatori di una ditta siano sempre attentissimi e pronti anche sul fronte della sicurezza.
    Sto dicendo delle ovvietà, ma è per capire meglio i vostri interventi.
    Grazie mille!

  6. #186
    ●⁞◌ Ȏrȉzzȏntέ Ðέglȋ ȨvέntȊ ◌⁞●
    Registrato
    Aug 2008
    Località
    Palermo
    Messaggi
    2,952

    Predefinito

    ripeto il concetto espresso nel mio primo intervento: bisogna prestare attenzione ai sistemi operativi su base
    open modificati direttamente dai vari produttori
    di dispositivi per aggiungere delle funzionalità, in quanto
    senza potercene rendere conto, una serie di rischi alla base di tali modifiche, sono sempre alla nostra porta.

    Quindi il mio è un atteggiamento quantomeno scettico in merito alle attività del team di sviluppo Synology
    che dopo aver creato il problema è anche intervenuto in ritardo, solo successivamente alla scoperta della
    grave falla di sicurezza da parte di un ricercatore indipendente.

    Già in precedenza una serie di utenti avevano mostrato sul forum ufficiale dei dubbi e chiesto spiegazioni in
    merito a traffico sospetto individuato dalla continua (ed immotivata) accensione dei led di stato del traffico
    dei loro NAS in assenza di attività proprie. Ma la causa è stata scovata solo esternamente al Team.



  7. #187

    Predefinito

    Originariamente inviato da squonk
    In merito alla discussione su Synology volevo cercare di capire meglio il senso dei vostri interventi.
    Totocellux ha indicato che i programmatori di Synology si sono mossi tardi rispetto a delle vulnerabilità note da tempo. Hanno poi tappato le falle con gli aggiornamenti.
    La necessità di tenere un sistema aggiornato all'ultima versione è (per me) sempre fuori discussione.
    Ma permettetemi di chiedervi di chiarire meglio il senso di questi discorsi in questa sede.
    Stiamo dicendo che il software di Synology è più sicuro di altri perchè ha una casa produttrice che ci sta dietro? Quindi sarebbe da preferire un progetto come XPEnology rispetto ad altri progetti open come FreeNAS e OMV?
    Vuol dire che progetti come gli ultimi citati non sono così sensibili alle questioni di sicurezza e vengono aggiornati con meno frequenza?
    Io ho sempre pensato il contrario...
    Sulla bontà del software Synology ho visto poche discussioni: in molti, quando si sceglie un NAS commerciale, optano per quest'ultimo per la bontà del DSM, ma non è detto che gli sviluppatori di una ditta siano sempre attentissimi e pronti anche sul fronte della sicurezza.
    Sto dicendo delle ovvietà, ma è per capire meglio i vostri interventi.
    Grazie mille!
    E' la stessa cosa per la maggior parte dei sistemi: di sistemi open (come OMV e FreeNAS) ne esistono molti e hanno una fetta di utenza minore = difficilmente gli hacker perdono tempo a scovare falle..sistemi più diffusi come DSM, sistemi di altri NAS, Windows, firmware di router di marche famose ecc.. sono usati da più utenti = gli attacchi da parte degli hacker aumentano ma hanno un team di supporto alle spalle a cui tocca mantenerlo sicuro rilasciando continui aggiornamenti. Synology fino ad'ora ha supportato ampiamente i suoi NAS e i relativi sistemi per quanto riguarda sicurezza e aggiornamenti.
    Spero di essermi spiegato bene

  8. #188
    kibibyte
    Registrato
    May 2012
    Messaggi
    311

    Predefinito

    Finalmente il mio microserver è attivo e funzionante!!!
    Ho deciso per OMV che ho installato su un disco SSD da 40gb (costo circa di 30€ spedito) e attualmente ho un unico disco WD Red da 4Tb.
    Mi sono impratichito con la configurazione facendo dei test su una macchina virtuale, poi ho settato la macchina reale.
    Sono stato a lungo indeciso sul filesystem con cui formattare il disco dei dati, in particolare la scelta era fra EXT4 e XFS, ma alla fine ho optato per il rodato EXT4.

    L'unica cosa su cui ancora molto incerto è il settaggio del risparmio energetico sui dischi fisici (Advanced Power Management) di cui ho capito poco.
    C'è qualche esperto di Linux che potrebbe aiutarmi a capire?

  9. #189
    Super Moderatore L'avatar di frakka
    Registrato
    May 2001
    Località
    Casalecchio di Reno (Bo)
    Età
    44
    Messaggi
    23,405
    configurazione

    Predefinito

    Io avrei lasciato XFS ma effettivamente credo sia un problema marginale.
    Per l'apm che problema c'è? Non ci ho mai perso molto tempo ma ci possiamo guardare...

    Questa è la storia di 4 persone chiamate Ognuno, Qualcuno, Ciascuno e Nessuno. C'era un lavoro importante da fare e Ognuno era sicuro che Qualcuno lo avrebbe fatto. Ciascuno poteva farlo, ma Nessuno lo fece, Qualcuno si arrabbiò perché era il lavoro di Ognuno. Ognuno pensò che Ciascuno potesse farlo, ma Ognuno capì che Nessuno l'avrebbe fatto. Finì che Ognuno incolpò Qualcuno perché Nessuno fece ciò che Ciascuno avrebbe potuto fare.

  10. #190
    kibibyte
    Registrato
    May 2012
    Messaggi
    311

    Predefinito

    Sulla scelta del filesystem ho letto diversi articoli e visto diversi benchmark, ma molti sono relativi a situazioni che non riguarderebbero il mio uso (vedi l'interazione con database, ecc.).
    Ho letto di utenti che consgliavano XFS in particolare per file molto grandi (salvo poi trovare un articolo in cui dice espressamente che con file molto grandi va in crisi) ma non ho trovato indicazioni o bench per un insime di file variegato, cioè con dimensioni molto variabili: io ho diverse foto da 2M l'una, diversi video da 350M, musica in flac con files dai 30M in su e qualche file mkv da 8 o 9 Gb...
    Alla fine nella grande indecisione ho scelto EXT4 perchè più diffuso e anche perchè da quello che ho visto le differenze nelle prestazioni non sono poi così eclatanti.
    Ma dico questo: se mi dai una buona ragione per passare a XFS, formatto adesso finchè il disco è ancora quasi vuoto!

    Riguardo al Power Management.
    Per ogni disco OMV propone i seguenti parametri da settare (credo siano settaggi comuni in Linux):
    Advanced Power Management
    -- Disabled
    -- 1 – Minimum power usage with standby (spindown)
    -- 64 – Intermediate power usage with standby
    -- 127 – Intermediate power usage with standby
    -- 128 – Minimum power usage with out standby (no spindown)
    -- 192 – Intermediate power usage with out standby
    -- 254 – Maximum performance, maximum power usage
    Automatic Acoustic Management
    -- Minimum performance, Minimum acoustic output
    -- Maximum performance, Maximum acoustic output
    Spindown Time
    -- 0 to 360 minutes
    Write Cache
    -- Enable write-cache. (This function is effective only if the hard drive supports it.)
    Capisco poco le differenze fra i vari valori del primo gruppo e anche a cosa serve il valore "Spindown Time".
    Inoltre non capisco questi valori come influiscono sui dischi normali e sugli SSD.
    Ovviamente vorrei una configurazione che propende verso il risparmio energetico e sulla silenziosità (penso a tutte le ore nella giornata in cui il NAS rimarrà acceso senza essere utilizzato), sia per risparmiare corrente sia per preservare l'hardware il più possibile.
    E' ovvio che non vorrei poi soffrire troppo sul fronte delle prestazioni.

    Al momento ho optato per i seguenti settaggi sul WD (ma senza capirne a fondo il significato):
    Advanced Power Management
    -- 127 – Intermediate power usage with standby
    Automatic Acoustic Management
    -- Minimum performance, Minimum acoustic output
    Spindown Time
    -- 30 minutes
    Write Cache
    -- Enabled
    Sul disco SSD credo che certi valori non abbiano valenza. Ho settato:
    Advanced Power Management
    -- 64 – Intermediate power usage with standby
    Automatic Acoustic Management
    -- Disabled
    Spindown Time
    -- Disabled
    Write Cache
    -- Enabled
    Ogni suggerimento e chiarimento è davvero gradito!

Pagina 19 di 23
prima
... 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 ultimo

Informazioni Thread

Users Browsing this Thread

Ci sono attualmente 1 utenti che stanno visualizzando questa discussione. (0 utenti e 1 ospiti)

Regole d'invio

  • Non puoi inserire discussioni
  • Non puoi inserire repliche
  • Non puoi inserire allegati
  • Non puoi modificare i tuoi messaggi
  •  
nexthardware.com - © 2002-2022